Copyright (c) 2001 The Harvard Law Review Association
Harvard Law Review
November, 2001
115 Harv. L. Rev. 170
LENGTH: 62788 words
COMMENT: EROG .V HSUB AND ITS DISGUISES: FREEING BUSH V. GORE FROM ITS HALL OF MIRRORS+
+ The terror that befell America on September 11, 2001 has led not only to great sadness but also to the reevaluation of much that might initially seem quite distant from the ruins of the World Trade Center. Even though this Comment was essentially completed by the time of the attack, I might have had second thoughts about the wisdom of publishing it now if its point were to question the legitimacy of the Bush presidency. I understand the view that wartime conditions make the legitimate authority of the Commander-in-Chief all the more relevant a topic of inquiry, and I would defend the patriotism, as well as the right, of anyone who voices dissent while others struggle to bring unity to the nation (and indeed to the world) in its war on terrorism. The right to dissent is part of what the terrorists who murdered thousands of innocent children, women, and men would like to destroy. But the right to dissent is not a duty. To feel solidarity with the nation's head of state at such a time is not in the least shameful. So I understand and sympathize with the position taken by The New York Times as reported by its chief political correspondent, Richard L. Berke, twelve days after the attack: "In a move that might have stoked the partisan tensions - but now seems utterly irrelevant - a consortium of news organizations, including The New York Times, had been scheduled this week to release the results of its ambitious undertaking to recount the Florida presidential ballots. (That has been put on hold indefinitely.)" Richard L. Berke, It's Not a Time for Party, but for How Long?, N.Y. Times, Sept. 23, 2001, 4, at 3. The point of this Comment, however, is not in the least to question the legitimacy of the President's claim to office. When Al Gore conceded defeat, leading to the selection of George W. Bush by the electoral college as confirmed by Congress, any cloud that might have been cast by the extraordinary proceedings that transpired between Election Day on November 7 and the U.S. Supreme Court's decision day on December 12, 2000 ceased to have relevance to the formal legitimacy of the presidential succession. Indeed, I think those proceedings soon ceased to have much relevance to this President's legitimacy as head of the executive branch and Commander-in-Chief of the armed forces even in an informal political and cultural sense. To those who would portray this President as the corrupt choice of a conservative cabal, to those whose mission it is to prove that the Bush v. Gore majority acted out of character and thus inadvertently revealed their partisan motives, I leave the challenge of deciding whether their cause is worth pursuing at this moment in our history. Far from discrediting the Justices in the Bush v. Gore majority for acting out of character, this Comment seeks to discredit them for acting so much in character that they provided an unusually clear window into much that is lamentable about how they discharge their role: with utter disdain for democracy and its pluralistic institutions and with eyes fixed firmly on surface appearances rather than underlying realities. When terror strikes where all of us live, the last thing we should do is relax our vigil against assaults on liberty and law, however well-meaning, by those who have forgotten, or never learned, the meaning of democracy. Believing this Comment to be part of that vigil, I decided that the awful events of this September gave me more reason, not less, to publish it and indeed even to leave untouched the jocular way in which it begins.
* Tyler Professor of Constitutional Law, Harvard Law School. I am grateful to Bruce Ackerman, Mike Dorf, Heather Gerken, Tom Goldstein, Pat Gudridge, Jonathan Massey, and Joel Perwin for helpful suggestions and to Sanket Bulsara, Dan Geyser, Mike Gottlieb, Tara Grove, Latonia Haney, Ben Hatch, Dori Kornfeld, Steve Lehotsky, Justin Levitt, Eric Miller, Angie Nguyen, Corey Stoughton, Hien Tran, and Kevin Walsh for excellent research assistance and editorial advice. All errors are, as the saying goes, my own.
SUMMARY:
... Perhaps, in the press of time, the Justices were unable to agree on an airtight justification for their decision. ... Only the concurring opinion of Chief Justice Rehnquist, joined by Justices Scalia and Thomas, rejected the Florida Supreme Court's reading of the state election code's definition of a "legal vote," on the basis of an "independent" interpretation that gave "deference" to the state's highest court but only within bounds of reasonableness that the concurring opinion concluded that court had exceeded. ... Yet Justices Ginsburg and Souter, both of whom were quite exercised in Bush v. Gore about federal judicial intervention bearing on the allocation of authority between the state legislature and the state judiciary, joined Justice O'Connor's majority opinion in Rogers without distancing themselves from any part of it. ... The real question is whether the degree of accuracy programmed into the machines is, in extremely close elections, a degree of accuracy with which the Florida Legislature would have been content. ... Undaunted by the logical flaws in their analysis of what might constitute a "legal vote," the state court's critics try to buttress their argument by claiming that their interpretation accords with that of the election officials upon whom the state legislature conferred the primary duty of construing and applying Florida's election laws. ...
TEXT:
[*172]
"Honk if you've heard enough about Bush v. Gore!" Not an implausible bumper sticker - even before the decision's first anniversary. [*173] If you supported Bush, you may well be thinking, "Get over it! It was a one-shot decision that ended a miserable partisan mess that could've gone who-knows-where. The Court's critics have had their day. Now that the Court has taken a bullet for the country, it's time to move on." Even if you supported Gore, you may well be thinking, "Enough already. Now that the Court's most strident critics - the 673 law professors who denounced the Court right after it announced its stay, n1 and the well-known authors who claimed they could prove its motives were partisan and nefarious n2 - have had their air time, why pile still more attacks on something that's behind us? We'll have our turn in 2004!"
It's hard not to think that the air has been all but sucked out of whatever room may have existed for sober reflection about this notorious judicial decree. Despite the collections of thoughtful essays that have appeared n3 and will continue to appear n4 in print, it seems likely that - except for diehard opponents of Bush who will never come to terms with the Court's lightning-like intervention - the majority of Americans will embrace a standard story that will put the entire matter to rest.
That story might go something like this:
Once upon a time, there was a presidential election between a compassionate conservative candidate and a liberal candidate unsuccessfully marketed as a dynamic populist. The election proved to be so close that its outcome turned on the twenty-five electoral votes of one state, Florida, where the margin of error probably exceeded the margin of victory. The rules in place on election day would have given a narrow but certain victory to the conservative. But the loser had the good fortune of belonging to the political party that had appointed all seven members of the state's highest court, a court that - upon request [*174] from the losing candidate - read new terms into the state election code while reading old terms out and gave several county canvassing boards - all hand picked by the loser with only partisan gains in mind - its blessing to conduct manual recounts of ballots in a dubious effort to "divine the intent" of the voters. These ballots had already been counted twice, any untabulated ballots were the result of voters' not having followed directions, and any further counting was likely to damage both the physical integrity of the ballots and the integrity of the election itself. But the liberal canvassing boards - eventually joined, at the Florida Supreme Court's invitation, by their counterparts across the state - counted and recounted under wildly divergent standards from county to county and even within counties themselves. As the count droned on, the conservative's margin of victory grew alarmingly narrow, so narrow that unprincipled partisans seemed on the verge of overturning the people's democratic choice. So the U.S. Supreme Court, fully cognizant that it would be excoriated in the liberal press and legal academy, soberly intervened to stop the recount and restore order only hours before midnight on December 12, 2000, the moment when Florida would have lost its chance to participate fully in the electoral college. Perhaps, in the press of time, the Justices were unable to agree on an airtight justification for their decision. But whatever one thinks of the legal technicalities of the various opinions, surely we should all be grateful that the Court prevented a constitutional crisis that might have imperiled our democracy.
However popular that tale might be, it seems unlikely to displace entirely another story, this one quite different in its orientation. This second story might replace the pro-Supreme Court story, coexist with it as an equally compelling narrative, or lurk in the background as a bitter anti-tale that would go something like this:
The only reason that too-close-to-call election was ever close in the first place is that the network-anointed winner, a true-blue conservative successfully marketed as a moderate, had the entire political machinery of Florida on his side. The posterboy of modern nepotism had his brother calling the shots from the governor's mansion and the co-chair of his Florida campaign as the state official in charge of certifying the election results. What's more, he imported a veritable army of his and his father's cronies from out of state to file lawsuits "making a federal case out of it" in an attempt to protect his lead by stopping any recounts that could help his nemesis, and to storm ballot-recounting sites in an attempt to prevent by force what they could not stop by law. Witnessing a spectacle that seemed calculated to frustrate democracy, the Florida Supreme Court did its best to reconcile a confusing election statute when called to duty by the parties. It may have had to fudge the letter of the law a bit, stretching a deadline here or a recount standard there, but it complied with the purpose of the statute in a justifiable effort to make each vote count.
[*175] It was the U.S. Supreme Court, in this alternative story, that sacrificed principle to partisan motive. It saw the handwriting on the wall - that counting every vote could well mean a victory for the liberal and thus an administration that would pack the Court with judges hostile to its staunch conservative agenda. Driven by this self-aggrandizing motive, the five conservatives who comprised the Court's majority agreed to stop the recount, departing from their long-held states' rights principles and from their previously narrow construction of the Equal Protection Clause. After stopping the recount, the Court held a perfunctory oral argument and, by a narrow majority that split along strictly political lines, conveniently decided that no time remained to complete a recount that could comply with their constitutional rule du jour - a rule cobbled together from the doctrinal remnants of the very Courts these conservatives had spent their careers attempting to bury in the graveyard of legal history. There can thus be no explanation but gross and unprincipled partisanship for the Court's order to anoint Bush II President of the United States on December 12, 2000.
It seems worthwhile to deconstruct both of these fairy tales as decisively as possible, lest either of them ever be cemented into the national psyche as conventional wisdom. Doing that, and replacing both caricatures with a more balanced account of the Supreme Court's role in the presidential election of 2000, is the threefold aim of this essay.
The more balanced account goes something like this:
The presidential election of 2000 came down to the wire: everything depended on determining who had won Florida's twenty-five electoral votes. The state first attempted to figure that out by conducting an automatic statewide recount on the day after the election, which indicated that Governor Bush had narrowly taken the state. Pressure mounted for Vice President Gore to concede and let the country move on with Bush as its leader. But news reports revealing widespread election-related wrongdoing in Florida gave Gore pause. The state's election code mandated a recount in all sixty-seven of its counties, but the required recount was performed in just forty-nine of those counties. Many Florida voters felt disenfranchised by the so-called "butterfly ballot" in Palm Beach County, which they said had misled them into voting for Pat Buchanan rather than Al Gore for President. Nothing could be done about that injustice because there was no legal way to rerun part of a presidential election. But something could be done about the large number of people who had tried to vote for one of the presidential candidates but had cast ballots that the voting machines in their counties had treated as containing no votes at all for President. Manual recounts could at least uncover those so-called "undervotes," which according to one reasonable reading of the Florida election code, constituted legal votes, provided the voter's intent could be clearly ascertained.
[*176] Even if Gore had wanted to concede, taking that step rather than trying to count those votes would have betrayed the tens of millions of voters across America who had voted for him - more, as it turned out, than had voted for Bush. Following Florida's election code, Gore's lawyers picked four heavily Democratic counties in which to lodge protests of the tallies, triggering manual recounts (something Bush's lawyers could have done in heavily Republican counties but chose not to do).
Four counties began to conduct manual recounts - only to be told by Secretary of State Katherine Harris that the results would be excluded from the final tally unless the recounts were completed within a week of the election. Gore and Bush supporters battled in the state courts over the legality of the Secretary's action, and the case went to the state supreme court - all of whose members had been appointed by Democrats. That court ruled on November 21 that the ambiguous (and seemingly inconsistent) provisions of the election statutes, when interpreted as a whole and in light of the fundamental place of the franchise in the state constitution, required Harris to give the counties five more days to complete their recounts. The Bush lawyers challenged that conclusion, and the state court's underlying interpretation of state election law, in the U.S. Supreme Court. They argued that the recounts should be stopped at once because they violated Article II of the Constitution, which directs the state legislatures to fix the method of choosing each state's electors, and 3 U.S.C. 5, which offers states what I have termed a "safe harbor," insulating their chosen electors from congressional challenge, provided all controversies surrounding their selection are resolved, in accord with rules and procedures in place on election day, by a date that worked out to be December 12, 2000. The Supreme Court refused to stop the recounts but unanimously sent the case back to the state's highest court for a better explanation of how its ruling complied with Article II.
Bush simultaneously challenged the recounts in federal court, seeking to enjoin the whole recounting process on the ground that it was so discretionary and variable that it deprived Bush of due process and equal protection, but the federal courts dismissed his action, finding no basis to circumvent the state courts, which could decide all the federal constitutional claims.
The saga dragged on. Secretary Harris certified a slate of twenty-five Bush electors on November 26, and Gore filed what state law calls a "contest" to challenge the election result on the basis that enough legal votes had been rejected by the voting machines to "change or place in doubt the outcome of the election." n5 The state supreme [*177] court, this time dividing 4-3, accepted Gore's contest and on December 8 launched a statewide manual recount of all the undervotes.
Some wondered if the nation's fate would have to be decided by one state's highest court. Such fears were unfounded. The Constitution's Twelfth Amendment, aided by several federal statutes, provides the procedures for Congress to act as the final arbiter of a contested presidential election.
That holding was based on a skewed picture of the situation in Florida. As is true in most states, there were enormous differences among counties in voting machines, types and designs of ballots, and other variables that powerfully affected the odds that a voter's intentions would be accurately tallied. These underlying inehe whole case was a no-bra_Út oualities might have existed among counties with respect to methods of recounting ballots. Yet the Court refused to see beyond the surface inequalities in the recount and insisted that a clear set of objective rules, uniform across the state, was needed to solve the alleged constitutional problems.
But, alas, said the Court, there was too little time for the state court to formulate such rules because its own earlier opinions had supposedly said that state law mandated completing the recount process by December 12. And in two short hours it would be December 13. Q.E.D.
The Chief Justice, with Justices Scalia and Kennedy, not only joined the per curiam, but also concurred separately to urge an alternative ground of decision. According to the concurrence, the state supreme court did not construe the Florida election code in accord with the rules enacted by the legislature pursuant to its Article II authority: it was "absurd" to construe the state's election code to treat ballots cast in violation of the instructions as though they contained "legal votes" [*178] just because one could clearly discern what the voter had intended to communicate. Voters who failed to follow the rules had nobody to blame but themselves if the voting machine spat out their ballots. On that view, the whole case was a no-brainer, because there could an imagined Article II issôWhat e" if the machines were functioning properly; thus the entire recount was a lawless exercise. And the concurrence added, for good measure, that the state legislature "must have" intended that there be no recounts past December 12.
That this more balanced account describes the reality of Election 2000 more fairly and accurately than either caricature is part of what I hope to persuade readers of all political stripes. My intent is to dispel the suspicion that Florida's highest court played fast and loose with the state's election statutes, while showing that the U.S. Supreme Court acted in a manner wholly inconsistent with its constitutional responsibilities, whether viewed in terms of equal protection and due process or in terms of Article II; that it had no warrant to interfere with the political process as it did; but that its having done so was sadly of a piece with much that the Court has done in recent years.
After setting the stage in Part I with a description of the lawsuits that culminated in Bush v. Gore, I pause in Part II to show that the entire Article II issue that has so fascinated many commentators is a red herring and that observers on both sides have conjured an Article II problem where none really exists. In Part III, I show that the one question giving rise to an imagined Article II issue - whether the Florida Sstitution that has animate_‰d ering rather than construing the state's election code - is easily resolved: far from changing, breaking, or even bending anything in state law, the state's highest court interpreted that law in a manner fully consistent with its letter as well as its spirit. In Part IV, I show that the equal protection and due process attacks on the recount are completely without merit, looking for a genuine pea under each of a series of shells: a one-person, one-vote shell, a structural equal protection shell, and a due process shell. Finding no pea, I examine the possible grounds for the persistent intuition that there may nonetheless have been something unfair about the recount and come up empty again. In Part V, I show that the entire challenge to the recount presented a political question for resolution by Congress rather than a justiciable question for an Article III court. In Part VI, I show that none of my critique of the Supreme Court's holding presumes partisan motives and that what the Court did is perfectly understandable in terms of several unfortunate pathologies generally manifested in its jurisprudence. Finally, after commenting on the exceptionality of Bush v. Gore in Part VII, I draw some lessons for the future in Part VIII, concluding that the real challenge is not to strip the Justices of their robes but to expose and erase the flawed vision of our Constitution that has animated so many of their recent r we all learned of the inmÆfinee work of democracy in a way that can give the [*179] lie to that vision, inspiring the successors of today's Justices, if not them, to join in a more noble constitutional venture.
You might wonder how someone with an ax to grind n6 could be an appropriate choice to undertake that project in a measured way. There is nothing I could possibly say now that could answer that question. The proof of that pudding will have to be in the eating.
I. Election 2000: The Grinch That Almost Stole Christmas n7
"Great cases, like hard cases, make bad law. For great cases are called great, not by reason of their real importance in shaping the law of the future, but because of some accident of immediate overwhelming interest which appeals to the feelings and distorts the judgment. These immediate interests exercise a kind of hydraulic pressure which makes what previously was clear seem doubtful, and before which even well settled principles of law will bend." n8
What I find most interesting about Bush v. Gore n9 isn't just how wrong its legal reasoning seems but what shaped the Court's perceptions and how its performance fit into the patterns of decision that are coming to define the Rehnquist Court. Right after we all learned of the infamous "butterfly ballot" total immersion gave me až_e baNew York Times editorial n10 urging patience while Florida's legal processes determined who was disenfranchised and what to do about it. I thought we would "have an opportunity, as Dec. 18 approached, to think about just what to do if the irresistible force of litigation meets the unmovable obstacle of time." n11 Little did I suspect that the institution that would prove unable to resist the "force of litigation" would be the U.S. Supreme Court, or that I'd find myself plunked in the middle of the whole affair, writing the briefs as Vice President Gore's Supreme Court counsel and presenting his case in the first of the two Supreme Court oral arguments. I can't pretend I didn't get a kick out of being at the center of attention, but there were gnawing questions: "Should this be the center? Are people like me supposed to be playing the lead [*180] roles in this show? Should people like them - the nine Justices questioning me - be calling the shots here?" Dozens of Senators and Representatives were sitting in the public gallery listening to the dialogue. "Shouldn't they be more than spectators to this drama?"
Bush is the only case I have ever studied in real time, teaching it and beginning to write about it at the very moment I was litigating it. Though I therefore have little basis for comparison, I believe my total immersion gave me a clearer view of the way Bidly growing sense that noú< onest, became Bush v. Gore, the lawsuit, which in turn captivated the nation as a televised spectacle that forever changed the image, and ultimately the reality, of all it touched.
In trying to get our bearings, it may be helpful to begin with the situation that led to and surrounded Bush v. Gore, the saga that began with the first premature announcement of Gore's having won Florida's crucial twenty-five electoral votes on Tuesday evening, November 7, 2000. That faded into the "never mind" mea culpas from the media as they announced in the predawn hours of November 8 that the winner was in fact Bush. That in turn blurred into the embarrassed explanations shortly thereafter that the election in Florida, and thus in the whole nation, was technically too close to call. A pitched public relations battle ensued, with Bush attempting to depict Gore as a sore loser and Gore reciting the mantra "every vote counts." Gore's efforts were boosted by the sight of elderly Jewish voters in Palm Beach, many of whom would as soon vote for Pat Buchanan as buy a ticket to the moon, whom the butterfly ballot had apparently misled into voting for Buchanan when they meant to vote for Gore. Meanwhile, televised pictures of the butterfly ballot made clear to millions of Americans how easy it had been to lose one's franchise in Florida. The rapidly growing sense that nothing could be done about closed, finally, with a maŠça laough it doubtless cost Gore thousands of votes, more than enough to make up the Bush margin of victory) underscored the urgency and, for many, the justice of letting Gore hunt for uncounted legal votes wherever he could find them in Florida. n12 But the candidates' spinmeisters continued to spin as the televised reality lingered lastingly and mockingly over the painstaking effort to recount ballots manually in counties where Gore and his supporters had lodged protests. The television-constructed image of people trying to see through seemingly opaque ballots made it easy to deride the on-again, off-again recount as a thinly veiled attempt to change the election's rules at the end of the game. "Dimples" and "hanging chads" became household words that evoked close-up images of election fraud for some, while [*181] conjuring memories of the struggles over voting rights in the 1960s for others.
The television cameras zoomed in as talking heads gravely intoned predictions of "constitutional crisis" at the thought that Congress might be drawn into the process - unless we could be delivered from that supposed evil, and from what many evidently perceived as the dangers of democracy, by the deus ex machina of the Supreme Court, described by journalists in hushed tones suggestive of a latter-day Wizard of Oz. The saga closed, finally, with a made-for-TV ending that featdnight on December 12, even whilorters, clutching the Court's "final answer" as they dashed down its marble steps, the imposing white facade of the Supreme Court illuminated by banks of floodlights in the crisp December night. Less than two hours before midnight on December 12, even while network anchors struggled - live ae counting of ballots thenP-e haourt's deliberately obscure bottom line, one thing was clear: the Court had ended the recount, bringing the presidential election of 2000 to a suitably dramatic, if dramatically undemocratic, close.
Flashback: The case that ended as Bush v. Gore in the U.S. Supreme Court began with a lawsuit that George W. Bush filed in a federal district court in Florida four days after the election of November 7 had freakishly failed to yield a winner. Bush sought to remove that rather awkward obstacle to his assumption of the presidency by obtaining an injunction against the continued counting, under state judicial supervision, of disputed ballots in the struggle over Florida's decisive block of twenty-five electoral votes. In that federal suit, which seems already to have vanished from most memories of the episode, n13 Bush claimed that principles of judicial federalism, which would ordinarily counsel abstention from such federal intrusion into the ongoing processes of a state's judicial system, should give way here in light of the federal constitutional principles that would supposedly be sacrificed were the counting to proceed.
In essence, those principles were these: because the people of Florida, as Bush construed the election code that their legislature had enacted, had not authorized the counting of ballots then underway, the state courtngoing legal processes to __ fed were guilty of undemocratically displacing the people's choice with judicial fiat; circumventing the directives of the state legislature regarding the selection of the state's presidential electors, in violation of Article II, Section 1, Clause [*182] 2 of the U.S. Constitution; n14 depriving the state's duly chosen slate of electors of the security against congressional challenge offered by the 1887 safe harbor statute, 3 U.S.C. 5; n15 permitting essentially indistinguishable ballots to be counted differently depending on where in the state they happened to be cast and when in the counting process they happened to be tallied, in defiance of the one-person, one-vote principle and in a discretionary manner inviting the injection of partisan bias; and resolving the presidential election through a set of rules and principles that were not in effect when that election began.
Vice President Gore requested my assistance at that early stage, when the issue was whether the federal courts should provide injunctive relief on the basis of theories that then-Governor Bush could readily present to the courts of Florida, whose final ruling would be reviewable by the U.S. Supreme Court on certiorari. I agreed to argue the Vice President's case, believing freedom from this sort of federal interference in a state's ongoing legal processes to be an important part of ouember 4 to ask Florida's h¶óead d, even more so, of the constitutional pluralism I have long embraced. This principle echoed one I had pressed over a decade earlier in the Supreme Court on behalf of Pennzoil, when I argued that Texaco should not have been awarded a federal injunction against the enforcement of a Texas statute that required the company, after it had suffered an adverse multibillion dollar judgment in the state's courts, to post a bond in the full amount of that judgment if it wished to prevent the state court from executing the judgment against its assets pending appeal. n16 The Supreme Court agreed unanimously with the views I espoused, n17 and I thought it plain that the same arguments should prevail on behalf of Gore in Florida, particularly because I saw Article II's explicit delegation to each state [*183] of the responsibility to select a slate of presidential electors as an added shield against federal intrusion, not the sword the Governor sought to make of it.
In a case that would abound with ironies, the success of our arguments against Bush in the federal courts - we won at the trial court level n18 and on appeal to the Eleventh Circuit, n19 and the Supreme Court denied certiorari n20 - contributed two ironies of its own: First, Article II turned out to provide sword enough to lead a unanimous Supreme Court on December 4 to ask Florida's highest court to explain how its earlier extension of a state law ballot-counting deadline from November 21 to November 26 had derived from state legislation, as Article II required, while simultaneously providing the opportunity for the Chief Justice, joined by Justices Scalia and Thomas, concurring in the Court's December 12 decision to end the election, to conduct an independent analysis of state law and to conclude that the state court had unjustifiably departed from the legislative scheme. Second, the Vice President, seeking a tireless and talented trial lawyer to press in the Florida courts his case for counting all the ballots, turned to the lawyer who had enginee ballot cast by an eligiblû_ing cessful foray against Pennzoil into the federal courts: n21 David Boies. It was undoubtedly among the best moves Gore made in the post-campaign campaign: although I had prevailed over Boies in Pennzoil v. Texaco, n22 Boies's strategy at the lower court level in Pennzoil had been brilliant, and his success in the Florida courts proved to be at least as impressive. n23
What made that success so difficult was the novelty of the situation - there was no precedent for reexamining ballots cast but not counted for either candidate (so-called "undervotes") to resolve a statewide presidential election that was too close to call - coupled with the perverse [*184] audacity of the Bush assault. It was as though the Bush lawyers had the foresight to imagine the strongest possible arguments that could be made against the kind of Supreme Court victory they ultimately obtained, then reflected those arguments in a diabolical mirror capable of displaying what looked like legal propositions even though they were in fact nonsensical or at least logically inverted, and then put those pseudo-arguments - those [in' '] - forth, without embarrassment and indeed with genuine conviction, as arguments against the Florida courts' construction of the state's election code in a manner that favored counting every legal vote, that is, every ballot cast by an eligible voter that clearly manifested the voter's intent.
It would seem that a Supreme Court decision stopping the counting of such ballots at a point when election controversies could no longer be finally resolved by December 12 would contravene the process the Florida Legislature, as understood by its highest court, had set in motion; would undemocratically displace the people's choice with a judicial ukase; would deprive the state's electors of any hope of falling within the federal safe harbor; n24 would violate the one-person, one-vote principle by arbitrarily distinguishing between ballots counted for a candidate and those tossed out as not machine-readable despite the clarity of the voter's intent; and would do all this in the name of a federal constitutional ruling wholly without preelecting the United States<Ia Nual election of 2000 under principles not in effect on November 7.
Sound familiar? It should. Those became the very charges Bush leveled against the Florida Supreme Court for permitting the counting of these ballots. To attack the Florida court on that basis, the lawyers for Bush, to borrow an image the Supreme Court once used in another context, had to "step through an analytical looking glass" to devise their challenge. n25
II. The Red Herring of Article II
Rarely has a case been so much studied but so little understood, even by those who have looked at it most closely - whether as critics or as defenders. It is as if Bush v. Gore has come equipped with its own hall of mirrors, the better to project misleading images of what was at stake, leading students of the underlying litigation and the ultimate judicial resolution down false trails of inquiry.
Central to the ways in which observers have been misled has been the role, in resolving the election stalemate, of a hitherto little-noticed [*185] provision of the Constitution: Article II, Section 1, Clause 2, which provides that "each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors" charged with selecting the United States president. n26 Critics of which federal constitutioÎfoffin into Florida's processes for deciding which presidential electors to appoint have proceeded as though they had to attack the U.S. Supreme Court for disagreeing with the Florida Supreme Court's interpretation of Florida law, even though Article II's explicit delegation of authority to the Florida Legislature and not to the Florida Supreme Court seemingly authorized this disagreement. Similarly, defenders of the Court's intervention have proceeded as though they must defend the Court for intruding into the internal structure of Florida's government, finding in Article II the authority to reject a state constitution's allocation of authority between the state legislature and the state judiciary. Thus, many critics have found themselves all but reading Article II, Section 1, Clause 2 out of the Constitution, while numerous defenders have found themselves treating that provision as justifying quite radical inroads into a core feature of state sovereignty.
Both positions have been rendered especially awkward for those forced into them. For critics, an aversion to federal superintendence of processes by which the organs of state government impinge upon the selection of federal officials is difficult to reconcile with their own positions on state efforts to impose term limits on federal officers n27 or in other contexts in which federal constitutional norms restrict state aurt propounded, despite hi_×ridarace of federal superintendence of those processes is difficult to reconcile with what they have said in those same settings, treating state control over even the federal officials who represent the state's people as too central to state sovereignty to be limited by federal constraints. n29
[*186] Once we bring Bush v. Gore into proper focus, however, it should become clear that both the critics and the defenders of the Court's decision have been aiming at phantom targets. To take the critics first, when they argue that the U.S. Supreme Court had no business second-guessing the Florida Supreme Court on a question of Florida law notwithstanding Article II, they make the mistake of attacking the Court for something that, as it happens, it did not do in Bush v. Gore. They make the simultaneous mistake of needlessly claiming that Article II gives the Supreme Court no authority to do just that in appropriate circumstances. n30 They therefore doubly miss their mark, leaving their target unscathed once it is shown that the Court was innocent of the sin they attribute to it and leaving themselves wounded by the kickback effect of having fired a shot in a direction difficult to defend.
The Supreme Court did not in fact reject any reading of Florida law that the Florida Supreme Court propounded, despite hints in its initial remand me Court affirmatively disêwusedrred as a matter of law in concluding that Florida Secretary of State Katherine Harris abused her discretion in cutting off the time for manual ballot recounts in counties where Gore lodged protests. n31 Only the concurring opinion of Chief Justice Rehnquist, joined by Justices Scalia and Thomas, rejected the Florida Supreme Court's reading of the state election code's definition of a "legal vote," on the basis of an "independent" interpretation that gave "deference" to the state's highest court but only within bounds of reasonableness that the concurring opinion concluded that court had exceeded. n32
To be sure, the Court's per curiam opinion, while not disagreeing with the Florida Supreme Court's reading of the state's election code, nonetheless purported to read in the court's rulings a determination that the election code valued completing the recount in time to qualify [*187] for the federal safe harbor above counting all legal votes. Critics may properly fault the Court for having ventured that transparently speculative "reading" rather than remanding the case to the Florida Supreme Court with directions that it explicitly make that determination. n33 That criticism, which is taken up later in this Comment, n34 is a serious one indeed. It should not, however, be confused with a criticism that the Supreme Court affirmatively displaced the decision of a ssuppose the state's highes_mhe ping of state law.
But what if the Supreme Court had done just that? Apart from wasting their ammunition by shooting at a target not really in play, those who assail the Court for independently construing state legislation, and even for intruding into a matter of utmost import to the state's project of self-governance - namely, the state constitution's structuring of the internal relationships among the three principal organs of government - are shooting themselves in the foot. Of course [*188] the federal judiciary has a role to play in policing what a state's courts do with respect to the manner in which presidential electors are chosen. If a state court were to rule that only white males who own real property in the state may vote for presidential electors, would anyone doubt that the federal judiciary could properly intervene? True, the basis for intervention in that case would be the Fourteenth Amendment, but what of it? Suppose the state legislature, acting as authorized by Article II, Section 1, Clause 2, were to say that, on the date specified by Congress pursuant to Article II, Section 1, Clause 4, n35 there shall be a "popular election in which all qualified voters shall be eligible to participate, with the winners of a plurality of the votes cast being declared the presidential electors." And then suppose the state's highest court ruled: "Notwithstathe Ex Post Facto Clause, G ple.ain preference for a popular vote, it is the view of this court that the people are dunderheads and that this court should, and it hereby does, designate the presidential electors as follows ... ." Would anyone doubt that, in this case as well, the federal judiciary could properly intervene - indeed that it would be derelict if it did not?
Plainly, the federal judiciary must police more than just the Fourteenth Amendment when presidential electors are involved: it must ensure compliance with Article II and every other provision of the federal Constitution that in some way constrains the process for choosing presidential electors.
The governing constitutional proposition is straightforward and, in my view, not genuinely debatable: whenever a provision of the federal Constitution specifies something - whether substantive, structural, or procedural - about how a decision otherwise internal to a state's system of governance should be made, that provision's enforcement is a matter for the federal judiciary, and ultimately the Supreme Court, subject to the political question doctrine or other principles committing the matter to the federal political branches. The post-Bush v. Gore case of Rogers v. Tennessee n36 provides one particularly telling example. In Rogers, the Court construed the Ex Post Facto Clause, which provides that "no St[state] government." n38 The obvst facto Law." n37 The Clause's usage of the words "pass" and "Law" clearly indicated to the Court that the Clause was "a limitation upon the powers of the Legislature, [*189] and ... not ... the Judicial Branch of [state] government." n38 The obvious result of construing the Ex Poshe state "Legislature" is º›ly, rt did was for federal courts to constrict state legislatures more tightly than they constrict state courts in an important realm of decisionmaking that bears on individual rights, regardless of how a state might prefer to allocate authority between those two branches. Yet Justices Ginsburg and Souter, both of whom were quite exercised in Bush v. Gore about federal judicial intervention bearing on the allocation of authority between the state legislature and the state judiciary, n39 joined Justice O'Connor's majority opinion in Rogers without distancing themselves from any part of it. n40
Indeed, although Justice Ginsburg's Bush v. Gore dissent, joined in part by Justices Souter and Breyer, n41 expressed grave concern about the authority of the U.S. Supreme Court to dictate how a state arranges its internal organs of decision, n42 it seems worth stressing that even the concurrence of Chief Justice Rehnquist, which was the only opinion that came even close to supervising Florida's election process in such an intrusive way, in fact avoided actually doing so except in dictum. For the Rehnquist concurrence did not hold that Florida had violated Article II by assigning to its judicial rather than legislative branch the task of deciding the manner of choosing presidential electors. Plainly, the role Article II assigns to the state "Legislature" is fully consistent with any Article II), clearly misinZL1] constitutional rules not limited to the presidential context: rules governing the number and nature of the chambers of the state legislature; specifying whether state legislators are chosen from districts or at large; spelling out the legislature's internal voting rules; identifying the role of the state governor in vetoing legislative measures subject to legislative [*190] override; n43 or setting forth state constitutional norms. In this sense, the federal Constitution undoubtedly takes each state's legislature as it finds it. n44 But at least some state constitutional assignments of responsibility away from the state legislature - specifying that the manner of choosing presidential electors was to be fixed by the state's highest court, say, or by the state's chief executive, or by the people in a statewide plebiscite n45 - obviously would not be consistent with Article II. n46 Importantly, neither the Court nor the concurrence in Bush v. Gore hinted that Florida had made any such assignment.
The only suggestion that Florida violated Article II came in the Chief Justice's concurring opinion: it suggested that the Florida Supreme Court, discharging its interpretive function in full accord with state arrangements that left the final lawmaking power with the state [*191] legislature (as required by Article II), clearly misinterpreted the governing stentrenched" n52 in the staÂvther it described as "deference," the concurring opinion deemed the state court's interpretation so far outside the range of what was "reasonable" as to be "absurd." n47 I think that conclusion was itself utterly unreasonable. That, however, is a substantive disagreement, not a charge that the concurring opinion poked its nose into a realm that was none of the Supreme Court's business. For it plainly was the Supreme Court's business, as it always is when a federal constitutional norm speaks directly to a challenged exercise of state power. Indeed, despite intimations in the dissents of Justices Stevens, Souter, Ginsburg, and Breyer that the Supreme Court was meddling in a strictly state sphere and that the concurring Justices in particular were forgetting their states' rights stripes, n48 it would be difficult to sustain the view that any meddling or poaching was afoot. n49
In Rogers, for example, both Justice Ginsburg and Justice Souter joined an opinion that, having distinguished the norms applicable to state legislatures from those applicable to state courts, n50 proceeded to dissect prior rulings of a state's highest court to determine whether that court had correctly characterized its own precedent. n51 This dissection played an essential role in the Court's decision regarding whether a rule of law had been "firmly entrenched" n52 in the state's law prior to the chant has already been said sh½úthe defendant before it. n53 Plainly, then, Justices Ginsburg and Souter, the two Bush v. Gore dissenters in the Rogers majority, and Justice Breyer, a third Bush v. Gore dissenter who agreed with the Rogers majority's "basic approach," n54 would have to concede that a proper function for the Supreme [*192] Court is to police a state's highest court's interpretation of state law, whether judge-or legislature-made, n55 to ensure compliance with governing federal constitutional norms, unless the political question doctrine renders the case nonjusticiable.
Thus, those critics of Bush v. Gore who decry the decision, or even the concurring opinion, as an unauthorized federal sally into an exclusively state domain or as impermissible federal judicial second-guessing of a state's highest court on a question of state law, have been trying to tackle a phantom runner: such a sally, if there had been one, would not necessarily have been unauthorized, and some degree of second-guessing would have been entirely proper. None of this is meant to deny that the Rehnquist concurrence's particular interpretation trips all over itself; my point here is simply that the problem with that concurrence is not that it asked an inadmissible question, but that it gave an indefensible answer.
What has already been said should make plain why Bush v Supreme Court's interpret¦_ncurnecessary work for themselves when they assume that the Court's ruling n56 depends for its lawfulness upon a reading of Article II, Section 1, Clause 2 that is on a collision course with state sovereignty - the core of which, as the Court recognized in Gregory v. Ashcroft, n57 is the allocation of responsibility for governance among the three branches. n58 Dicta in the concurring opinion notwithstanding, nothing that the concurring justices did required an assumed federal judicial power to revise a state's assignment of responsibility over a category of decisions to its judicial branch rather than its legislative branch. Moreover, not even the dicta in that opinion suggest that any substantial issue for the Supreme Court to resolve is raised by a state constitution's conventional assignments of responsibility that round out the state legislature's role in directing a mode of selecting presidential electors - for example, the gubernatorial responsibility of vetoing legislation subject to override only by a supermajority, or the [*193] state judicial responsibility of construing legislation to resolve ambiguities, fill gaps, eliminate contradictions, and harmonize the whole with state constitutional protections for fundamental rights such as equality of the franchise.
The concurring opinion rejects the Florida Supreme Court's interpretation of the state electioate legislation may be tosÌ_ staurt's opinion that a vote not cast in full accord with the voting instructions is nonetheless a "legal vote," provided the voter's intent is clear. n59 Although the concurring opinion's application is unsound on the merits, the institutional function of checking the state court's construction of state election legislation to ensure that federal constitutional ground rules (here, those of Article II) are followed is unexceptional. Of course, it must be conceded that this checking function authorizes the Supreme Court to reject only manifestly unreasonable state judicial constructions of state statutes and not simply to substitute its own preferred construction for the state court's. Thus, in his dissenting opinion in Rogers v. Tennessee, n60 Justice Scalia - joined in this respect by Justice Thomas and Justice Stevens and thus speaking for two-thirds of the Rehnquist trio in Bush v. Gore - had no difficulty treating as "binding upon [the Court]" what he described merely as a "reasonable reading of state law by the State's highest court," n61 notwithstanding the fundamental federal right that he accused the Court's majority of insufficiently protecting. n62
It therefore seems futile, as well as unnecessary, for defenders of Bush v. Gore to talk as though the interpretive rulings of a state's highest court construing state legislation may be tossed aside without at least Even before theOsards federal constitutional norm that gives state legislatures a particular role is in the picture. n63 This is the case whether that norm deals with the rights of individuals accused of crime, as with the Ex Post Facto Clause, or with the responsibilities of choosing presidential electors, as with Article II, Section 1, Clause 2. Some degree of deference, within an outer perimeter whose definition is, of course, a federal question, is required notwithstanding the presence of a controlling federal norm. In any event, the concurring opinion in Bush v. Gore purported to pay such deference and proves vulnerable not because of how it postured itself vis-a-vis the [*194] state's highest court, but rather because its substantive reading of the state's election law, not to put too fine a point on it, turns out to be baloney. And it bears reiterating that the Court's per curiam opinion did not read the state's election law in any manner at odds with the reading that the state's highest court gave it. It was in the way the opinion read the Fourteenth Amendment, both in the equal protection theory that all but two Justices essentially endorsed and in the remedial ruling that five Justices embraced, that the Court stumbled - and stumbled badly.
III. Florida Election Law: The Law of Standards?
Even before they signed on to the per cur fatal attraction to the a“æa yehree Justices had already tripped up by endorsing the reasoning of Chief Justice Rehnquist's concurring opinion, which argued - as have other noteworthy defenders n64 of the outcome in Bush v. Gore - that the Supreme Court's intervention was needed to bring an end to a lawless process that the Florida Supreme Court injudiciously set in motion. Judge Richard A. Posner, building upon the argument of the concurrence, has painted the Court's decision in Bush v. Gore as a pragmatic response, perhaps not exactly lawful but not entirely unlawful either, reasonably designed to contain the damage that might otherwise have ensued from the Florida Supreme Court's unjustified and unreasonable decisions of November 21 (extending the deadline for certification) and December 8 (ordering a statewide manual recount of all undervotes n65). n66 Yet by rejecting an inflexible, "voters-must-go-by-the-book" approach to the Florida election law in favor of a construction that permitted the tabulation of as many intent-revealing ballots as possible, the Florida Supreme Court adhered faithfully to the primacy of the right to vote as articulated in the literal text of both the [*195] Florida election code and the Florida Constitution. n67 By contrast, critics of that court's reasoning - apparently driven less by legal analysis than by a yearning for finality and a nearly fatal attraction to the allure of surface precision Florida Supreme Court rejmýi0 scheme they purported to uphold.
A. Round 1: The Extension
The Florida Supreme Court initially demonstrated its unwillingness to read the state's election code as though its recount provisions were designed to trap the unwary during the protest phase of the Election 2000 litigation, which followed the decisions by four counties to grant the Florida Democratic Party's November 9 requests for manual recounts. n68 Soon realizing that they might not complete the vote tabulation process by November 14 (the statutory deadline for submitting certifications of election results), three county canvassing boards inquired of Florida Secretary of State Katherine Harris in what circumstances she would accept late returns. n69 Her response: "Almost never." In accordance with a Division of Elections advisory opinion, n70 Harris reconciled the oft-quoted (and admittedly contradictory) "shall" and "may" language in the Florida election code n71 as supporting her asserted ability to reject late returns on the ground that a canvassing board's tardiness was seldom excused. Harris would grant a reprieve [*196] to counties only in cases of voter fraud, substantial noncompliance with election procedures, or "an act of God." n72
On November 21, the Florida Supreme Court rejected Harris's determinati provide a rationale for cá_ch sd in effect chosen to teach tardy canvassing boards a lesson at the cost of the votes of tens of thousands of Florida citizens. It held that "the Secretary may reject a Board's amended returns only if the returns are submitted so late that their inclusion will preclude a candidate from contesting the certification or preclude Florida's voters from participating fully in the federal electoral process." n73 As will soon become clear, Harris's interpretation was utterly inconsistent with the unambiguous preference in Florida law for relaxed standards which permit, whenever possible, the counting of every lawfully cast ballot. n74 Yet, to many Court-watchers who heard only television snippets about how the state supreme court had managed to construe a conflict between "may" and "shall" reject late returns to mean that the Secretary "must" accept such returns, the state court's decision hardly seemed to reflect a logical reading of existing Florida rules. Instead, the decision seemed politically motivated and thus likely to aggravate the tension and uncertainty that many perceived to be taking hold in Florida.
The state court's proposed fix, to permit the recounts thwarted by executive officials' (mis)interpretation of the election code to continue until November 26, only exacerbated such sentiments. n75 The court did not provide a rationale for choosing that deadline unti The chapter of the ElectiÔ¬ full week after the U.S. Supreme Court had vacated and remanded its decision, hours after conclusion of oral arguments and submission of the case, n76 and barely more than a day before the U.S. Supreme Court rendered its final ruling. Florida's highest court finally explained on December 11 that counties would have had until November 18 to certify their final results even absent any protests or manual recounts. n77 The Division of Elections' misinterpretation of the state's election code on November 13 had deprived the counties of five days during the original time period to recount ballots manually and still have the resulting votes included in the certified total. Thus the Florida Supreme Court on November 21, in extending the deadline to November 26, simply gave back those five days. n78
But the state court's explanation came too late to placate many who from the beginning had viewed its November 21 ruling as ad hoc judicial legislation. Eyebrows were raised and subsequent rulings from that same court would meet deep distrust from those who saw Florida's highest court as flouting the aphorism that the "rule of law" is the "law of rules." n79
[*198]
B. Round 2: The Statewide Recount Order
The chapter of the Election 2000 saga that would ulillegal votes or rejection of a urt's final word in Bush v. Gore began on November 27, 2000, when Vice President Gore filed suit under a provision of the Florida election code that permits an election contest in cases involving the "receipt of a number of illegal votes or rejection of a number of legal votes sufficient to ent of the voter as determźõere ult of the election." n80 Florida's highest court determined that Vice President Gore had indeed made a threshold showing that the tabulating equipment had rejected enough "legal votes" to "place in doubt" the results of the presidential election. n81
The court then turned to the question of remedy. Reasonably inferring from the broad grant of authority in chapter 102.168(8) n82 a mandate that the court "do everything required by law to ensure that legal votes that have not been counted are included in the final election results," n83 the court expanded its focus beyond the votes that Gore had specifically identified (the votes that had triggered the court's remedial powers) to order a statewide manual recount n84 of all undervotes on which one could discern the "clear indication of the intent of the voter." n85
C. Damaged and Defective Ballots?
The Florida Supreme Court, in formulating its "intent of the voter" standard, relied on chapter 101.5614 of the Florida election code, which regulates the canvassing of election returns. n86 This provision [*199] mandates that "all valid votes shall be tallied by the canvassing board" n87 and commands that "no vote shall be declared invalid or void if there is a clear indication of the intent of the voter as determined by the canvassing boawith Florida Supreme Court _ry tvides that "the return printed by the automatic tabulating equipment, to which has been added the return of write-in, absentee, and manually counted votes, shall constitute the official return of the election." n89
One recurring text-based criticism of the Florida Supreme Court's reading of the statutory language, which appears to have originated in the December 8 dissent by Florida Chief Justice Wells, n90 is that chapter 101.5614(5) is altogether inapplicable when determining a general definition of "legal vote" because that chapter applies only to the particular circumstance in which a "ballot card ... is damaged or defective so that it cannot properly be counted by the automatic tabulating equipment." n91 Chief Justice Rehnquist adopted this objection wholesale without any independent analysis of the statutory text. n92 Judge [*200] Posner likewise contends that "the provision about recording a vote when there is a clear indication of the voter's intent is for cases in which the ballot is damaged or defective, which is different from its being spoiled by the voter, and is indeed a kind of tabulating error." n93
With all due respect to its distinguished sources, this criticism is far wide of the mark. It not only strains the statutory text but also conflicts squarely with Florida Supreme Court precedent that authoritat asked with obvious irritaèýing 14(5) two years before Election 2000. In Beckstrom v. Volusia County Canvassing Board, n94 the Florida Supreme Court "construed "defective ballot' to include a ballot which is marked in a manner such that it cannot be read by a scanner." n95 This definition clearly encompasses incompletely punched-through ballots. n96 Although referring to Beckstrom for a different proposition elsewhere in his opinion, n97 Chief Justice Rehnquist conspicuously ignored the only portion of that prior state court opinion that directly addressed the meaning of "defective ballot" in chapter 101.5614(5). Judge Posner, for his part, dismisses Beckstrom by opining, in essence, that Florida's highest court could not really have meant what it said in 1998. His contention is a wholly unsupported assertion whose only virtue is that [*201] it permits him to advance his far more rigid interpretation with a straight face. n98
D. They Should've Read the Directions!
According to the analyses of Chief Justice Rehnquist and Judge Posner, no "legal votes" were missed if a machine counted all ballots without mechanical error, a tautological result under their circular definition of a "legal vote" as a ballot marked such that a vote-tallying machine can read it. n99 Echoing a question that Justice O'Connor asked with obvious irritation during the December 1er who, however innocentlyu_e amice Rehnquist argued that voters are, after all, expected to follow instructions:
No reasonable person would call it "an error in the vote tabulation' or a "rejection of legal votes' when electric or electromechanical equipment performs precisely in the manner designed, and fails to count those ballots that are not marked in the manner that these voting instructions explicitly and prominently specify. n101
[*202] The Chief Justice, however, failed to note that the instructions were not uniform from county to county and may have been provided only in English in areas with large Hispanic populations, despite the legal requirement that they also be printed in Spanish. n102
Critics of the state court decision seem frustrated by the fuss of having to deal with pesky votes that the voting machines could not adequately handle. Judge Posner, for instance, faults the Florida Supreme Court for failing, in its November 21 decision that extended the deadline for certification of manual recount totals in the protest phase, to appreciate "the bother of hand recounts - let alone the agony of hand recounts protracted beyond the seven-day statutory limit." n103 Judge Posner insists that "the question ... is the amount of inconvenience that a voter who, however innocently, has failed to follow direquires.
±_tranpose upon the election authorities ... ." n104
Chief Justice Rehnquist and Judge Posner seem to equate the statutory commands of the legislature with the programming commands of voting machine or software manufacturers and the physical limitations of the voting machines themselves. But voting machines are nothing more than translators that, within a certain degree of accuracy, process ballots into votes. The real question is whether the degree of accuracy programmed into the machines is, in extremely close elections, a degree of accuracy with which the Florida Legislature would have been content. The legislature explicitly required automatic machine recounts in very close elections; n105 allowed for county-based protests with on-demand sample manual recounts, and then, if an error in vote tabulation was detected in any county, full manual recounts of the votes in that county if such recounts would be likely to affect the outcome; n106 and flatly required, when ballots are counted manually, that "no vote shall be declared invalid or void if there is a clear indication of the intent of the voter as determined by the canvassing board." n107 Given [*203] these requirements, it seems most unlikely that a count free of machine error ipso facto provides the degree of accuracy in translation that the election code requires.
As thThe Florida __tionots into votes should remind us, ballots record votes but are not themselves votes. Counting ballots is therefore not equivalent to tabulating votes. While elsewhere Judge Posner recognizes the difference between ballots and votes, n108 he equates the two at crucial moments. For example, he accuses Gore's legal team of "equivocating skillfully between "counting a vote' and a vote's "counting,'" n109 and adds that the Gore team's formulation "made it sound as if there were thousands of ballots that had been cast by voters but never tabulated[, when i]n fact, all the ballots in question ... had been counted." n110 Chief Justice Rehnquist similarly stated that "no one claims that these ballots have not previously been tabulated; they were initially read by voting machines at the time of the election, and thereafter reread by virtue of Florida's automatic recount provision." n111 And therefore ... ? Whether the ballots were counted is obviously important, but just as obviously, not decisive. Under the Florida Legislature's scheme, counting all ballots by machine is a necessary but not sufficient condition to counting all legal votes; counting all the legal votes matters when a manual recount has been ordered because of a finding that there has been "an error in the vote tabulation which could affect the outcome of the election." n112
The Florida Supreme Court read the standerstanding. The defendern Buar manifestation of voter intent, rather than full technical compliance with all the stated rules, the critical test in determining which votes count as "legal" and thus which ballots must be counted. This interpretation of the state election code was consistent with that court's willingness to accept large numbers of absentee ballots from Republicans, despite the court's awareness that statutorily required information had been filled in by Republican Party workers rather than by the voters themselves, in undeniable violation of statutory requirements. n113 If securing a victory for Gore had been the Florida Supreme Court's agenda, as its critics either openly assert or silently assume, then rigorously enforcing the state's election statutes with respect to all absentee ballots, in accord with their literal terms (just as Chief Justice [*204] Rehnquist and Judge Posner urge), would have offered a vastly simpler and far more certain way to go. n114 And if the best reading, indeed the only reasonable reading, of Florida's election code is that a voter has not cast a "legal vote" if she fails to do all that the code, and the instructions set forth pursuant to it, require, then how the Florida Supreme Court could permit the counting of all those noncomplying absentee ballots, without which Gore would have overtaken Bush by a healthy margin, defies understanding. The defenders of the Bush victory in Bhing about reviewing partiÑã205]rgotten or generously forgiven the inclusion of these noncomplying ballots in the tally for Bush.
E. Because the Executive Said So!
Undaunted by the logical flaws in their analysis of what might constitute a "legal vote," the state court's critics try to buttress their argument by claiming that their interpretation accords with that of the election officials upon whom the state legislature conferred the primary duty of construing and applying Florida's election laws. n115 Thus, the Secretary of State's determination that ballots were automatically illegal if cast in a way that rendered them unreadable by the vote tabulation machines n116 was, in their view, entitled to considerable deference. n117 But the statutory provision governing election contests says nothing that even suggests that deference to the Secretary of State or to a county canvassing board is required. Rather, the statute stipulates that, as a remedial matter, "the circuit judge to whom the contest is presented may fashion such orders as he or she deems necessary to ensure that each allegation in the complaint is investigated, examined, or checked, to prevent or correct any alleged wrong, and to provide any relief appropriate under such circumstances." n118 Indeed, the [*205] contest provisions say nothing about reviewing particular certification-relateres the evidentiary signifÞLry tate as though they were the rulings of an Administrative Procedure Act agency. n119 And as far as the canvassing board is concerned, it is telling that the contest chapters, far from suggesting that decisions by the Secretary or by a canvassing board are presumed correct and are to be deemed conclusive unless found erroneous on review, merely specify that "the canvassing board or election board shall be the proper party defendant." n120 In short, the statutory language offers no support for Judge Posner's assertion that "principles of administrative law required the contest court ... to defer to the canvassing boards, as the experts in counting votes, unless their decisions were unreasonable." n121 Posner invokes these "principles of administrative law" as if they were a sort of natural law - a brooding omnipresence in some technocratic world. Nowhere, however, does he point to any provision of state legislation to anchor his "reading" in the statutory scheme established by the Florida Legislature. n122 Nor, for that matter, does he cite Florida judicial precedent adopting any doctrine similar to the Chevron deference familiar in federal administrative law. n123
In any event, the conclusion that the Florida Supreme Court's decision renders the protest phase a "meaningless preliminary to the contest phase" simply ignores the evidentiary significance the court attaches ment.
øhe c county canvassing board determinations. This significance is particularly evident in the court's rejection of Gore's claim that the Palm Beach County Canvassing Board improperly failed to recover votes from 3300 manually inspected ballots. The court stated that "the circuit court in a contest proceeding does not have the obligation de novo to simply repeat an otherwise-proper manual count of the ballots ... [and] the Canvassing Board's actions may constitute evidence that a ballot does or does not qualify as a legal vote." n124 Satisfied that the Palm Beach County Canvassing [*206] Board's refusal to register any of the 3300 ballots as votes in its manual recount during the protest reflected a proper factual determination that none of those ballots revealed a "clear indication of voter intent," and agreeing that this was the correct legal test to apply in counting votes, the Florida Supreme Court did exactly what Judge Posner consistently accuses it of failing to do: it deferred to the Board in an area in which the Board had expertise - manually recounting ballots using the legal standard that the court deemed proper. The court's deference in this area is wholly consistent with its refusal to defer on legal questions concerning which well-settled hornbook law - not some Sunshine State innovation -- calls on the court to make an independent judgment.
Even witunt. n131 Applying the "suŤ*uld inistrative law framework of Judge Posner's argument, deference to the executive is inappropriate if the agency's interpretation, however "reasonable" it might be in the abstract, n125 is contrary to the statute's plain meaning. n126 Florida case law, particularly when read in light of the state's constitutional provisions, reveals that the narrow definition of "legal vote" provided by the Division of Elections cannot be reconciled with Florida law as it has existed since 1975.
F. Not the Law of Per Se Rules
Until 1975 Florida common law oscillated between demanding strict compliance with election law requirements and providing a more relaxed interpretation of the statutes. n127 In Boardman v. Esteva, n128 the Florida Supreme Court stated definitively that it would "recede" from the more rigid approach and henceforth read the statutes to regard as valid all votes cast in "substantial compliance" with state regulations. n129 [*207] Thus in Boardman, despite numerous voter errors (such as sixteen ballots on which the voter did not indicate a reason for voting absentee) n130 and "clerical errors or omissions" (such as the loss of many outer envelopes containing absentee voters' affidavits), the court ruled that the absentee ballots should be included in the total vote count. n131 Applying the "substantial compliance" rulefraud by election officials. n13disputes, Florida courts have refused to invalidate ballots on which voters or officials failed to follow the precise letter of the law n132 and have rejected votes in large numbers only in cases of overwhelming evidence of fraud by election officials. n133
." n137 Whatever e·he lSupreme Court's adoption of the "substantial compliance" approach accords with the primacy of the right to vote in the state constitution. Article I, Section I of the Florida Constitution states that "all political power is inherent in the people" (a clause Judge Posner sarcastically labels the "people power" provision). n134 Judge Posner identifies the court's reliance on the state constitution and the pre-election day case law that interpreted and applied the "people power" provision as an instance in which the court "was going beyond conventional statutory interpretation." n135 Nonsense. Judge Posner's skepticism reflects a subtle jurist's uncharacteristically simplistic approach to interpretation: he proceeds directly from the statement that Article I of the Florida Constitution does not contain the words "right to vote" to the conclusion that one should be "mystified ... by the court's statement that "the right to vote is the pre-eminent right in the Declaration of Rights of the Florida Constitution.'" n136 Suffice it to say that nothing in the U.S. Constitution mentions a "right to vote" in a presidential election either; that didn't prevent the U.S. Supreme Court from saying in its Bush v. Gore per curiam that "when the state legislature vests the right to vote for President in its people, the right to vote as the legislature has prescribed is fundamental." n137 Whatever else may be controversial aase law interpreting the ex#of losition is not. It follows that nothing should seem [*209] mystifying about the Florida courts' derivation of a right to vote from the state's constitution, either.
Even if one deems the Florida Supreme Court's plain meaning analysis inconclusive, under its implicit "constitutional avoidance" approach, that court's interpretation is nonetheless clearly preferable to that of the Secretary of State. The Florida Supreme Court drew on its case law interpreting the election code for the basic propositions that ""the primary consideration in an election contest is whether the will of the people has been effected'" n138 and that "the right to vote, though not inherent, is a constitutional right in this state." n139 From these propositions, the court concluded that the "fundamental purpose of [the] election laws" in Florida is "to facilitate and safeguard the right of each voter to express his or her will in the context of our representative democracy." n140 By reading the relevant statutes in light of this right and hewing to the constitutional avoidance maxim, the court did not replace the scheme established by the Florida Legislature. It simply employed a heuristic for determining what that scheme meant in light of the background constitutional principles operative at the time of legislative passage and its own case law interpreting the election code - case law wixacting standard made it d?7winniliar.
G. An Unprecedented Definition of Legal Vote?
Chief Justice Rehnquist, in arguing that the state court's definition of "legal vote" lacked support in Florida law, stressed that the "intent of the voter" standard had never before been employed in a recount. n141 His concurring opinion pointed out that the Florida Attorney General, a Gore supporter, "confirmed in oral argument that never before the present election had a manual recount been conducted on the basis of the contention that "undervotes' should have been examined to determine voter intent." n142 But in so concluding, the Chief Justice overlooked entirely a major change in Florida election law, n143 a change [*210] that the Florida Legislature mandated to cure what it evidently saw as a defect in the common law of election contests. As the Florida Supreme Court noted in Gore v. Harris, n144 prior to 1999 the Florida election code had not spelled out the specific grounds for contesting an election. n145 The common law, as developed by the Florida courts, had permitted election contests only when the plaintiff could demonstrate that, "but for the irregularity, or inaccuracy claimed, the result of the election would have been different, and he or she would have been the winner." n146 This extraordinarily exacting standard made it difficult for plaintiffs whcount, much less one with mõnvinals had wrongly rejected legal votes to convince a state court to order a recount on their behalf. For example, in Anderson v. Canvassing & Election Board, n147 the losing Democratic candidate for the Board of Elections contested the rejection of eighty-one absentee ballots. The trial court, in a decision affirmed on appeal, agreed that some rejected votes should have been included in the total but denied relief because "the number of votes improperly rejected would not have changed the results." n148 The Florida courts seemed willing to relax this stringent standard only when the plaintiff offered substantial evidence of fraud. n149
In 1999, the Florida Legislature explicitly eased the burden on plaintiffs by amending the contest provisions to permit a challenge to election results whenever the "rejection of a number of legal votes" might "change or place in doubt the result of the election." n150 Thus, the fact stressed by the Chief Justice - that no manual recount had ever been conducted using the "voter's intent" standard to determine legal votes - had no bearing on the validity of the Florida court's definition of "legal vote." Instead, the absence of prior recounts under [*211] this standard reflected the hoops through which plaintiffs had to jump prior to 1999 to convince a state court to order any recount, much less one with a particular definition of extremely close electionsf_refeange in law that the Florida court's decision to order a statewide recount on December 8 reflected was the one the Florida Legislature had mandated the prior year.
Judge Posner laments that, under the Florida Supreme Court's interpretation of the state election code, "hand recounts in close elections would become the rule rather than the exception," n151 a state of affairs he describes as "all upside down." n152 Upside down by whose compass? Certainly not that of the Florida Legislature. The election code indisputably mandates a full statewide recount, using whatever method (machine or manual) each county used in the first instance, whenever the margin of victory is less than one half of one percent, as it was here. n153 In addition, the election code allows for on-demand sample manual recounts during the protest phase whenever a candidate files a written request with the county canvassing board, n154 authorizes full countywide manual recounts when the initial manual recounts indicate that there has been "an error in the vote tabulation which could affect the outcome of the election," n155 and specifies a standard for use during manual recounts to determine whether a ballot contains a legal vote. n156 And viewing the matter ex ante, potential voters and candidates might very well prefer a regime of manual recounts in extremely close elections. That preference makes go are no doubt moved by the_]in tt unduly worried by the visible risks of human error or fudging in any manual recount, a risk that vigilant observers and other safeguards can at least minimize, but is properly wary of the invisible and more insidious risks of machine error or sabotage in machine counts. In any event, even with the 1999 change in the contest statute, there are likely to be relatively few hand recounts because the great majority of elections simply aren't that close. All of that is beside the point, however. Whether manual recounts become routine or remain rare, because their wisdom seems to enjoy the explicit endorsement of the Florida Legislature, therefore, Florida courts must treat them not as "all upside down," but as very much "right side up."
[*212]
H. A Voter's Intent: In the Eye of the Beholder?
The state court's critics' insistence that machine-generated results are always preferable to manual counts may evince a lack of faith in the reliability of human vote tabulators, but neither Chief Justice Rehnquist nor Judge Posner is so naive as to suppose that machines themselves are infallible. Something deeper must underlie their belief that no reasonable legislature would have left so much responsibility in the hands of human counters: they are no doubt moved by the recognition that a ballotal requirements n162 might"Xuntyct but instead is the conveyor of a distinctly partisan message - a message which, when examined by a human vote tabulator, may "look" different depending on that tabulator's personal political preferences. Machines may err, but such errors, even if abundant, are at least random. Human "errors," by contrast - whether deliberate or caused by the vote counter's unconscious desire to see a particular candidate prevail - are anything but.
Indeed, Judge Posner believes that the "manual recount [authorized on December 8] would have been both standardless and incomplete, and in the absence of judicial review ... tainted with partisanship." n157 That statement overlooks the specific provision for judicial review by a court with statewide jurisdiction in the Florida Supreme Court's recount order. n158 It also reveals a lack of faith in those citizens routinely designated by the Florida Legislature to count votes manually in regular elections. n159 It ignores as well two added statutory safeguards that Florida law provides to control partisan maneuvering in the manual counting process. The state election code requires recount teams to consist of members of both parties, n160 and it mandates that all recounts be open to public view. n161 That not every recount team in every [*213] county rigidly observed those procedural requirements n162 might justify a few raised eyeblls had closed and the ost¾ andenge, but it hardly warrants a categorical renunciation of the entire undertaking.
More importantly, using anxiety about partisanship at the recount stage as a justification for halting the recount and living with the results of the initial count implies that the initial count is itself above such reproach. Reality belies such a claim. Partisan calculations can permeate every choice made during the election process - including what type of ballot to use (who would doubt, for instance, that the failure to provide Spanish language ballots would, in many counties, hurt more Democratic than Republican candidates?); how long to keep particular polling places open; which voters to turn away for want of proper identification; which ones to turn back for suspicion that they have criminal records; whose requests for clarification of ambiguities at the polling place to treat seriously and whose to dismiss; and a host of other similar variables plainly subject to discretionary political manipulation both in the period preceding the election and on the day of the election itself. Indeed, the count that Secretary of State Harris certified on November 26 - the count made decisive by the Court's halting of the recount - itself contains a considerable number of votes recorded through manual counts and recounts conducted after the polls had closed and the ostensible election results hthe elimination of all opp@tof d63
Blissfully assuming that the initial count is not tainted by these political realities - and indeed thinking that one can draw a dividing line between decisions made before "election day" and decisions made after - in particular ignores the important role that human counters play in virtually all elections during the commonplace process of tabulating absentee ballots. n164 If the risk that ballot tabulators might let their partisan wishes affect their official tallies were sufficient to invalidate the use of visual/manual methods when translating ballots into votes, we would never include such absentee votes in a close election, given the sophisticated and current polling data available throughout the election process to those who manually count absentee ballots. n165 But excluding absentee ballots on account of that risk would, of course, entail the exclusion of overseas ballots, which in turn [*214] would prevent members of the armed forces from voting, in all likelihood harming Republican more than Democratic candidates, but in any event representing an unthinkable disenfranchisement. The point of dwelling on absentee ballots is not to defeat a proposal no one has made - the exclusion of absentee votes in all sufficiently close elections - but to expose the futility of demanding perfect neutrality and the elimination of all opportunities to make discretis the "cure," then the Co4°cati an election.
There is an even deeper problem in decrying manual recounts on the basis of a belief that the risks of partisanship in the counting process are unacceptably high in the absence of rigid counting rules. Beyond the often hidden subjectivity inherent in deciding which rigid rules to adopt, the denunciation itself is, at bottom, an outcry against entrusting political power to fallible human beings who might at any moment abuse it - an outcry, in other words, against democracy itself. Certainly, the Constitution puts constraints on the opportunities for elected or appointed government officials to gain unilateral control over elections and rig them for a favored candidate, but the existence of partisan motives, even zeal, cannot be deemed fatal to the vote-counting process when, as in the Florida election of 2000, well defined procedures exist to allow all candidates an equal opportunity not only to protest the criteria by which ballots are counted, n166 but also to monitor the process by which some ballots are recounted. n167 The involvement of human discretion and the ubiquitous potential for partisan influence on individual judgment cannot be deemed a defect to be cured, whether by the ostensibly impersonal agency of machines n168 or by the purposefully mechanical application of formal rules. For if that is the "cure," then the Constitution's basic scheme g boards in the respective2‡g ab to check ambition with ambition, n169 to balance the self-interest that tugs one way with self-interest that tugs the other, all the while seeking in the long run to instill virtues of civic republicanism - is the equivalent of surgery in septic conditions, more likely to harm the patient than to help, and the whole enterprise of popular government - government of, by, and for the people - is doomed.
[*215] Predicated on a very different premise, the election code of Florida attempts to harness rather than to exile partisan motives and political self-interest, while providing safeguards against partisan fervor. The principal safeguard rests in reliance on the integrity of the members of county canvassing boards, with public scrutiny helping keep those board members honest. For example, the protest system in Florida encourages each candidate to file protests in populous counties that lean heavily toward that candidate's party. n170 The state's election statutes simultaneously reveal the legislature's confidence that, even in the midst of partisan wrangling, closely watched election officials will perform their duties in an ethical manner. Indeed, in Boardman v. Esteva, the Florida Supreme Court opined that the election code, which vests "the canvassing of returns, including absentee ballots, ... in canvassing boards in the respective counties who make judgmenent of the voter" standard, as is," requires a presumption of proper performance. n171
The Florida Supreme Court's decision to reject uniform, mechanically applicable, statewide rules in favor of the legislature's "intent of the voter" standard, as implemented on a county by county or e recount would of course íGfor t a mindless or an underhanded choice of a standardless process. Rather, it is a considered and open decision which recognizes that, under the governing statutes requiring that ballots be counted in a public and visible manner, n172 county canvassing boards can and should be relied upon to apply the "intent of the voter" standard with integrity in manual recounts.
The state court's further decision to include partial recounts from the precontest period in the official election results, a determination vilified in the per curiam opinion and by supporters of the Court's halting of the statewide recount, n173 may simply have been a further attempt to implement legislative reliance on political competition instead of apolitical control. Critics of the state court's ruling overlook the incentive effects of such inclusion: a rule that permits inclusion of all legal votes identified through the preliminary manual recount, and only those votes, encourages each candidate to mobilize the county canvassing boards to count all votes in the precincts that the candidate deems most favorable. Including partial results thereby increases the likelihood [*216] that complete results will be obtained, a fact that a backward looking analysis, taking the U.S. Supreme Court's later intervention for granted, neglects. And a complete recount would of course serve the most fundamentaln image that seemingly droA9f a o preserve "the right to vote[, which] is the right to participate; [and] the right to speak, but more importantly the right to be heard." n174
Given the legislature's evident acceptance of judicially supervised partisan struggle as a way to facilitate democratic choice; given its explicit declaration that election officials are presumed to tabulate votes fairly in that partisan environment; given its unequivocal embrace in 1999 of an increasing number of election contests and, thus, of more manual recounts; and given its overwhelming preference for standards over rules in the election context, the vision that Chief Justice Rehnquist, echoed by Judge Posner, advances - whatever its virtues might be - would require the Florida court to "depart from the legislative scheme." n175
It is therefore clear that, far from being out of control - a state court whose wings a heroic U.S. Supreme Court had to clip - the Florida Supreme Court, by rejecting a formulaic, rule-bound interpretation of the state election code, was adhering faithfully to both the letter and the spirit of the statutory scheme that the Florida Legislature put in place and made applicable to presidential elections pursuant to its duty under Article II, Section 1, Clause 2. The image of a partisan state court run amok, an image that seemingly drove the Court's majority anherefore color what each cJsnd irs, n176 dissolves under close analysis, leaving in its place the straightforward picture of a workmanlike judicial tribunal doing its best, under trying and unprecedented circumstances, to apply an admittedly imperfect set of election rules. There is simply no reason to see Florida's highest court as some kind of lawless renegade bent on manipulating the rules to elect Gore to the presidency.
We are left with a puzzle: Why were the three Justices who joined in the concurring opinion, all of whom were persuaded that the recount should be halted at once on the ostensible ground that it flouted Florida's election statutes - and to a lesser degree the two others who thought that the recount complied with those statutes but violated equal protection principles in a manner impossible to repair in the time remaining and thus agreed to join in a per curiam ruling that ended the recount forthwith - so profoundly averse to the idea of manual recounts [*217] as a matter of course in close elections? Why their sense that no rational legislature would ever choose a system, notwithstanding what the Florida statutes seem to say, in which voters who failed to dot their i's and cross their t's might still get their votes counted - albeit through processes that are necessarily subjective and in which partisan motives might therefore color what each counter "sees"? Why the feeties awakened me to the drDðred san combat, with its close ties to rough-and-tumble popular democracy, must be cabined, buttoned down, rendered rule-like and predictable rather than embraced and celebrated? Perhaps the answer is the fear that democracy itself breeds disorder.
IV. The Emptiness of the Equal Protection Claim
There was a time when an aroused and active citizenry gathering political support for a favored cause or candidate did not seem terribly alarming to the Supreme Court. In 1986, when I represented a group of Berkeley activists agitating for rent control against landowners who felt threatened by that grass-roots effort, the Court, in Fisher v. City of Berkeley, n177 ruled 8-1 for my clients, rejecting the argument that such citizen-based action was preempted by the Sherman Act. n178 But barely a decade later the landscape had changed: In Timmons v. Twin Cities Area New Party, n179 I represented a minority political party challenging a state ban on "fusion" candidacies - candidacies in which two parties, usually a minor party and one of the two major parties, nominate the same candidate - that the major parties saw as threatening to their dominance. The Court's 6-3 decision upholding the state's power to ensure political stability n180 by preserving the shared hegemony of the two dominant parties awakened me to the drumbeat sounded by a cohesid practices at each pollin£', ac Rehnquist and Justice O'Connor, members of the Fisher majority, joined by Justices Scalia, Kennedy, and Thomas, supported quite often by Justice Breyer (as in Timmons) and at times by Justice Souter as well, n181 had by the 1990s coalesced into an alliance whose signature concerns were the avoidance of political ferment and fragmentation, n182 the preservation of established hierarchy, and the imposition, whenever possible, of formally realizable, mechanical rules instead of contextual standards [*218] requiring discretion and subjectivity in their application. n183 I had already decided to make these increasingly evident judicial preoccupations - and their subtext, a perplexing distrust of democracy n184 - a central theme of my advanced constitutional law course when, serendipitously, the electoral contest between Bush and Gore landed in court, soon to become Bush v. Gore.
It is, of course, a standard feature of elections in this country - including elections for President - that counties, and precincts within those counties, administer elections with considerable autonomy. As a result, variations from county to county are enormous. There are dramatic differences among the counties' choices of voting machines, ballot designs, manual or other methods for tallying the vote, access to polling places, rules and practices at each polling location, and absentee bing equipment makes doing śYut aa voter will not have his or her preferences accurately recorded are considerably higher in some places than in others. Counting variances tend to correlate closely with race and class. Partly for that reason, many states have been able to reduce perceived and actual inequity by manually recounting ballots in particularly close elections.
Rarely before the election of 2000 - maybe never - has this commonplace procedure been viewed on television, much less televised worldwide around the clock. Yet such was the fate of the presidential election conducted in Florida. Pursuant to a decision by the state's highest court on December 8, a statewide recount was to take place in all sixty-seven of the state's counties. Given the huge variations among different counties' voting equipment and procedures, it should have come as no surprise that good faith efforts to interpret ballots that looked alike but were cast in different counties did not yield identical results. n185 A punch-card ballot with the removable square next to a candidate's name only partially perforated and still hanging from the back of the ballot - a so-called "hanging chad" - might reasonably be understood one way in a county whose voting equipment makes it easy for just about any voter to punch cleanly through a ballot, but another way in a county whose voting equipment makes doing so a challenge for many vot seems to have persuaded _7the entee ballots, of course, vary in innumerable ways, n186 making it a virtual certainty that [*219] look-alike ballots in those categories would be interpreted at least somewhat differently depending on which county they happened to come from. All these variables fit into different state schemes that champion the ideal of decentralized and locally administered election procedures. That the operation of one such system was witnessed in living color during the Florida cliffhanger of 2000 simply exposed the raw edges of democracy in action - edges of a picture that, in earlier times, would have been celebrated as depicting a vibrant and well-functioning democracy going about the oh-so-normal but ever-so-imperfect business of translating millions of disconnected voices into the will of a collective people. Yet in this election, the Court saw the imperfections of our democratic machinery not as an opportunity, but first as a threat, n187 and then as a violation of the Fourteenth Amendment's Due Process and Equal Protection Clauses.
It was not a coherent constitutional vision or a body of constitutional doctrine, but rather the spectacle of individuals trying to "divine the intent of the voter" n188 by holding ballots up to the light and peering intently - when the presidency of the United States was at stake - that seems to have persuaded a Supreme Court majority tula would reduce variabiliÅ` it anded something better. n189 True enough, our society jails people for years, sometimes even executes them, on the basis of how a given jury interprets and applies a general standard like "proof beyond a reasonable doubt," or on the basis of how a particular jury infers the intent of the accused from pieces of circumstantial evidence at least as susceptible to different readings as are the marks on a ballot. But, said the Court, those situations differ from one in which a physical object - a ballot - is at issue, and in which a state court is in theory capable of replacing the general standard enunciated by a state legislature with a standard of its own creation, specific enough to be applied uniformly and objectively across the state. n190
[*220] Although a ballot is indeed a physical object, until the voting process is fully computerized, there will remain as many ways to fill in a ballot as there are ways of leaving a trail of circumstantial evidence in a civil or criminal case. Thus, the Court's reminder that the ballot is a thing does nothing to explain, much less justify, its equal protection holding. That a more mechanical formula for translating ballots into votes would reduce variability in how the state records look-alike ballots is true almost by definition, but it does not follow that such a formula would reduce variability in how the state treatsing to see what light this®´li0significant? Because the Fourteenth Amendment protects voters, not ballots. n191 Moreover, voters in a statewide election are unlikely to be concerned with being treated the same way as others who have the same intent they do; they are likely to care about being treated in a way that accurately and fairly translates their intent into a vote for the candidate of their choice. Thus, a formula that automatically discards every third ballot might rate a high score on the sameness scale but would do rather badly on the recording-actual-intent scale. The example is relevant because any mechanical formula, even if not rigged to disregard voter intent, is nonetheless bound to discard a number of ballots from voters who in fact had a clear intent that a manual count would have detected.
Thus neither the "thinghood" of the ballot, nor the potential for uniformity and for the minimization of human involvement and thus human judgment in the translation of ballots into votes, would appear to support the Court's conclusion that equal protection and due process principles demand a different, more uniform approach than the traditional one, as American as keylime pie, of letting each county make its own best effort, using fallible human counters, to get each voter's intent right.
Before attempting to see what light this picture can shed on the theoretical constructs that might be built to contain the Court's intriguing holding, it will be helpful to consider each of the several domains of constitutional doctrine that might conceivably furnish a framework for what the Court held. Because the Court drew explicitly on the one-person, one-vote line of cases, and because that body of doctrine alone shows at least some promise of providing support for a formulaic approach to voting rights, I begin by examining this justification.
When we discover that the pea we are looking for is not to be found under that shell, we turn to a nimization of human involvement and thus human judgment in the translation of ballots into votes, would appear to support the Court's conclusion that equal protection and due process principles demand a different, more uniform approach than the traditional one, as American as keylime pie, of letting each county make its own best effort, using fallible human counters, to get each voter's intent right.
Before attempting to see what light this picture can shed on the theoretical constructs that might be built to contain the Court's intriguing holding, it will be helpful to consider each of the several domains of constitutional doctrine that might conceivably furnish a framework for what the Court held. Because the Court drew explicitly on the one-person, one-vote line of cases, and because that body of doctrine alone shows at least some promise of providing support for a formulaic approach to voting rights, I begin by examining this justification.
When we discover that the pea we are looking for is not to be found under that shell, we turn to a slightly more complex possibility - that there might exist a new form of equal protection doctrine altogether, [*221] one that focuses not on the rights of individuals to treatment of one sort or another but on the rights and wrongs of alternative structures for democratic decisionmaking per se. Such a new form of dFirst and Fourteenth Amendments. I will have to conclude that, unfortunately, the pea isn't hidden under that shell either.
At the end of that constitutional shell game, we will be back where we began, with the conclusion that Bush v. Gore is about nothing beyond the simple intuition that Florida's procedures for deciphering ballots, common though they may be, just won't do because they are too - well, just too "unequal," much as Emperor Joseph II chided Mozart's work for having "too many notes." n193 Because I want to believe that more might lie beneath that reaction, a reaction seemingly more aesthetic than constitutional in character, I will address the possible sources of the Court's unease about the "intent of the voter" standard under which similar looking ballots might be interpnation.
imes and places. In the end we will see that such unease either is unwarranted or, if warranted, would point not to ending the recount but to invalidating the entire election, certainly in Florida and perhaps throughout the nation.
A. Speaking Theoreticaconstitutional reason justȨs" tame
In December of 2000, soon after the Supreme Court announced its decision, some Democratic voters in Florida began replacing their red-white-and-blue Gore-Lieberman buttons with black buttons with white lettering stating simply: "Equal Protection, My Ass!" This outburst [*222] has intuitive appeal, capturing the "where the hell did that come from?" reaction of many voters, lawyers, and academics who read the opinion. How could so flawed and peculiar an equal protection claim have prevailed? The idea that all voters should be treated with equal dignity is not controversial; but the notion that a state must develop and apply "specific standards" for construing the marks and indentations on machine-rejected ballots to guarantee "nonarbitrary treatment of voters" and "to secure the fundamental right" to equal treatment n194 comes from nowhere and overlooks several critical dimensions of the situation. Might some hidden logic nonetheless explain the Court's insistence on uniform, objective criteria for processing ballots in this situation? The Court's opinion appeared to draw upon not one, but three different types of equal protection-due process (hereinafter, "equal protection") analysis and, in effect, asked the voters to follow each of the three equal protection "shells" to try to find a hidden "pea" of constitutional reason justply: "Equal Protection, My Ass!" This outburst [*222] has intuitive appeal, capturing the "where the hell did that come from?" reaction of many voters, lawyers, and academics who read the opinion. How could so flawed and peculiar an equal protection claim have prevailed? The idea that all voters should be treated with equal dignity is not controversial; but the notion that a state must develop and apply "specific standards" for construing the marks and indentations on machine-rejected ballots to guarantee "nonarbitrary treatment of voters" and "to secure the fundamental right" to equal treatment n194 comes from nowhere and overlooks several critical dimensions of the situation. Might some hidden logic nonetheless explain the Court's insistence on uniform, objective criteria for processing ballots in this situation? The Court's opinion appeared to draw upon not one, but three different types of equal protection-due process (hereinafter, "equal protection") analysis and, in effect, asked the voters to follow each of the three equal protection "shells" to try to find a hidden "pea" of constitutional reason justifying the outcome.
Mesmerized by the Court's prestidigitation, voters might miss the fact that the pea has already been palmed: there is no hidden constitutional reason to uncover. None of these shells contains a defensible rationale, but each is sufficiently distracting to leave at least some observers thinking that there must be something valid hidden there.
1. First Shell: "One Person, One Vote". - Perhaps Bush v. Gore was nothing more than a traditional one-person, one-vote holding applied in a novel setting. n195 Perhaps support for a mechanistic mode of vote counting may be derived from the mathematical precision of Reynolds v. Sims's n196 one-person, one-vote formulation. But no one doubts that, in the Florida recount procedure that the Court found unconstitutional, each vote counts, and is counted, equally. Nor would that procedure result in differential weighting across districts - for example, by placing more people in one district, and fewer in another, so that a vote in the less populous district [*223] would go further towards electing a representative than would a vote in the more populous district. n197 For Florida's was an at-large election to choose a single slate of presidential electors: there was only one pool for all the voters, and only one pool for all the ballots being counted.
Following McPherson v. Blacker, n198 Florida could have arranged for its representatives to the electoral college to be selected on a district-by-district basis. Article II vests "the appointment and mode of appointment of electors ... exclusively [in] the states," n199 and Florida could certainly have done something unconventional with its voting districts if it had wanted. However, as the Bush Court recognized at the beginning of its analysis, once the Florida Legislature gave its voters the right to choose the state's electors, it had to do so in a non-discriminatory manner. n200 Accordingly, Reynolds would require, at the least, that any districts used to select Florida's members of the electoral college be equipopulous. The problem is that Florida didn't arrange its election in this way. Florida's at-large scheme in no way implicated the Reynolds skewed district concern.
More to the point, the voting rights jurisprudence of the Equal Protection Clause provides little succor to the Court. Reynolds and its progeny have become increasingly sensitive to the political dimensions of the voting process, finding in the rough and tumble of party politicking over the composition of voting districts a set of "neutral" justifications to contrast with the "racial" justifications found suspect in much recent voting rights litigation. n201 No one suggests, of course, that the process of tabulating votes should be subject to partisan manipulation. But "one person, one vote" is not the place to find an antidote to partisanship in a set of neutral principles or policies that would make possible the sort of mechanistic, formula-driven methods of vote tabulation required by the Court in Bush v. Gore.
On the contrary, Reynolds, aided and abetted by Davis v. Bandemer, n202 permits just the sort of partisan politicking the Bush Court seemingly wanted to exclude. While a degree of precision is required, n203 that precision is bought at the cost of permitting the parties [*224] to draw district lines that take political parties into account. n204 Given that Davis permits a great deal of explicitly political and partisan discretion in drawing the districts in which votes are counted, it seems odd that the Court would invoke the Reynolds line of cases in an effort to exclude political discretion altogether during the counting phase. No one doubts that the Reynolds line would prevent a state from adopting a system in which those who tally machine-rejected ballots manually are instructed to toss out ballots with ambiguous marks indicating an intent to vote for Bush but to count all the votes for Gore. But the Court neither could nor did base its repudiation of the Florida recount order on any such basis. Reynolds certainly would not condone any such scheme. It does, however, at least by extrapolation, permit a state to develop its own decentralized systems of vote tabulation for statewide recounts supervised by a single judge, no less than for countywide elections that are not overseen by one person (and are thus more, not less, vulnerable to unresolved disparities). The only requirement is that everyone's vote must be counted equally, within a margin of error. n205 Put bluntly, Reynolds clearly supports some small degree of inaccuracy in the count so long as the method of counting is fair, and Davis contemplates a large dollop of politics in developing the method of counting. The procedure that the Florida Supreme Court developed to implement the enactments of the Florida Legislature - a procedure that included representatives of the candidates and was overseen by an impartial magistrate - certainly passed this test. n206
[*225] To its credit, the Court did attempt to bolster its dubious equal protection reasoning with references to case law. The Court cited Gray v. Sanders n207 for the proposition that the Equal Protection Clause forbids "arbitrary and disparate treatment of voters in [a state's] different counties." n208 The Court also quoted Moore v. Ogilvie n209 for the proposition that granting "one group ... greater voting strength than another is hostile to the one man, one vote basis of our representative government." n210 The cases the Court chose to enlist, however, simply do not support its cause. They are immediately distinguishable on the ground that each case involved a plan implemented by the state legislature that clearly had the purpose and effect of granting greater voting power to a particular class - in both cases, rural voters. Gray and Moore, in fact, are so far from where the Court wanted to go in Bush, that the Court was forced to mischaracterize the cases just to make them seem relevant. Gray did involve disparate treatment of urban and rural residents, but it did not involve arbitrary treatment. Indeed, the state legislature had in place a careful plan which had the effect of diluting the votes of urban residents. In Moore, while the Court did describe the law at issue as "arbitrary," n211 it also found that it "discriminates against the residents of the populous counties of the State in favor of rural sections." n212 Thus in each case the law at issue targeted a specific group - urban voters - for disparate treatment. The problem with Bush v. Gore, and what distinguishes it from Gray and Moore, is that Florida did not single out any class for disparate treatment, an element usually a prerequisite to an equal protection claim. n213 Nothing in the record indicated that the Florida Legislature, the state judiciary, or the county recount teams intended to discriminate against [*226] any class, suspect or otherwise, including any "class of one." n214 Given the Court's inability to find a group or individual victimized by discrimination, its hand wringing about granting one group greater voting power than another is really beside the point. Did the Florida Legislature know when enacting the state's voting laws, or did the Florida Supreme Court know when construing them to dictate an "intent of the voter" standard, that certain groups were more prone than others to an elusive expression of intent when voting? An affirmative answer seems most unlikely, but even if we assume the contrary, the worst charge that we can lodge against Florida is an imperfect attempt to equalize the ability of voters to have their ballots counted. Equalizing voting power is, of course, what Gray and Moore were all about, and thus these cases would dictate the opposite outcome from that reached by the Bush Court.
2. Second Shell: Structural Equal Protection. - The alternative preferred by Professor Pamela S. Karlan is that the pea has been shuffled into the newly minted "structural equal protection" shell that she finds implicit in the line of cases arising out of Shaw v. Reno. n215 Traditionally, there have been three ways in which a voter might be injured: failure to weigh each vote equally ("one person, one vote"), vote dilution, and vote denial. n216 A court would have to look at an apportionment plan as a whole and compare the plan's various districts to determine whether one group of voters had been discriminated against by actually having their votes "count" less than those of other voters. Shaw changed all that. It rejected the customary statewide focus, and the concomitant comparison of one group of voters with another. n217 Instead, when faced with a Shaw claim, a court must consider the challenged voting district in isolation, n218 look to the individual rather [*227] than the group, n219 and determine whether the apportionment plan "rationally [could not] be understood as anything other than an effort to separate voters into different districts on the basis of race." n220 If so, the plaintiff is injured simply by having his race correlated with his potential vote. n221 Whether his vote really is underweighted, diluted, or denied n222 is immaterial.
Professor Karlan suggests that in both Shaw and Bush v. Gore the Court deployed a new, "structural" model of equal protection doctrine that "regulates the institutional arrangements within which politics is conducted." n223 This new structural model, though supposedly drawing upon the Equal Protection Clause to defend individuals' voting rights, in fact permits the Court to use the Fourteenth Amendment to extend its power to adjudicate political controversies n224 at the same time as it [*228] undercuts the authority of legislatures. n225 For Karlan, both in Shaw and in Bush v. Gore, individual voters' rights are simply a cover for the Court's structural ambitions, and the crack in the shell that reveals this pea is the parties' lack of standing: in the voting rights cases, because it was not obvious how Shaw-plaintiffs were injured; and in Bush v. Gore, because there was, in her view, no injury to claim. n226
Now, while it is undoubtedly true that the sort of standing that the Court conferred in Shaw is, from a traditional perspective, problematic, n227 some of the Shaw plaintiffs were injured in a fairly conventional sense and did have traditional standing. These individuals are the voters Professor John Hart Ely calls "filler people" n228 - nonminority voters placed in a majority-minority district either to negate their voting power by submerging them in districts where they will predictably be outvoted, or simply to ensure that the district's total population complies with Reynolds, n229 or both. Even accepting the Court's [*229] new focus on the challenged voting district rather than the reapportionment plan as a whole, it seems reasonable to argue that "filler people" suffer a legal injury much like that recognized in a traditional vote-dilution claim. n230 And whatever one might say about the standing of the non-filler plaintiffs in Shaw and its progeny, it is hardly credible - indeed, it borders on the fantastic - to argue that Bush himself lacked standing to press an equal protection claim in the Florida lawsuit. n231
Although, as Professor Karlan points out, Bush was not a registered voter in any of the challenged counties in Florida (and consequently could not assert Shaw-like standing), n232 he surely had third-party standing. n233 [*230] His injury was obvious: the Florida Supreme Court was in essence forcing the Governor of Texas to exchange twenty-five electoral votes the Florida Secretary of State had certified as his on November 26 - a number sufficient to make him the next president - for a recount to be conducted by a process he regarded as an unconstitutional roulette game rigged in favor of his opponent. n234 Bush potentially had standing on this theory to represent at least those who had voted for him and whose votes stood to be devalued during a recount. n235 More generally, he shared with all the voters a sufficiently common interest in protecting the integrity of the vote count to ensure his standing as third-party plaintiff. His standing was analogous to that of the defendant in Powers v. Ohio, n236 in which the Court allowed a white criminal defendant to represent black voters excluded from the jury pool on the basis of their shared interest in eliminating racial discrimination from the courtroom. n237 Because of the significant obstacles any individual voter would face in seeking to ensure a fair count, n238 Bush not only had standing but was particularly well placed to assert voters' equal protection [*231] rights - even though his ultimate goal was to have some of their votes excluded. n239
But if Bush had standing to sue based on individual harms suffered by some voters rather than based on a hypothesized harm suffered by the polity as a whole, then Professor Karlan's structural equal protection account of the decision becomes deeply problematic. Moreover, structural reasoning of the kind she identifies is goal oriented rather than value-neutral: it is directed at bringing about some form of change in legal or political processes, some improvement either intrinsic to the structure of governance that the doctrine requires or derived instrumentally from that structure. Structural equal protection arguments, like structural due process arguments, n240 do not occur in a vacuum, but are intended to create, entrench, or shake up a pattern of social change - for example, to alter an institution's or agency's decisionmaking process to ensure that it becomes more accountable or transparent, or more permeable to changing norms. n241
Witherspoon v. Illinois n242 provides an example of permeability-enhancing structural due process. In Witherspoon, the Court held that excluding individuals from serving on the jury in a capital case because they harbored doubts about the death penalty was unconstitutional. n243 The Court observed that "one of the most important functions [*232] any jury can perform ... is to maintain a link between contemporary community values and the penal system - a link without which the determination of punishment could hardly reflect "the evolving standards of decency that mark the progress of a maturing society.'" n244
Illustrating the accountability-enhancing version of structural due process, the Court in Hampton v. Mow Sun Wong n245 upheld a challenge by resident alien plaintiffs to the Civil Service Commission's blanket rule against the hiring of foreigners in federal civil service positions. The Court began with the proposition that "the business of the Civil Service Commission [is] to adopt and enforce regulations which will best promote the efficiency of the federal civil service," n246 not to exercise responsibility over foreign affairs. n247 Yet the Commission had not made a "considered evaluation" of the need for such a per se rule, failing to exercise "some degree of expertise, and ... [to] make known the reasons" for its conclusion. n248 Accordingly, the Commission's rule deprived resident aliens of liberty without due process of law. Similarly, in Regents of the University of California v. Bakke, n249 Justice Powell, writing only for himself but delivering the controlling opinion, expressed reluctance, on structural grounds, to permit the State of California to delegate to a state medical school's relatively unaccountable governing board the authority to ameliorate past racial discrimination, or to achieve educational diversity, by setting aside a number of places for qualified minorities. Justice Powell was concerned that "isolated segments of our vast governmental structures are not competent to make those decisions, at least in the absence of legislative mandates and legislatively determined criteria." n250 And in Ferguson v. City of Charleston, n251 a case decided several months after Bush v. Gore, the Court emphasized what it understood to be the pivotal role a local police [*233] department had played in shaping a municipal hospital's policy of testing for and reporting cocaine use by pregnant women for the "direct and primary purpose" of "using the threat of arrest and prosecution in order to force women into treatment." n252 Given the role of the police in formulating the policy, the Court concluded that the hospital violated the patients' Fourth Amendment rights when it "undertook to obtain such evidence from their patients for the specific purpose of incriminating those patients" without their knowledge or consent. n253
Invocations of Fourteenth Amendment rights to justify structural or systemic, rather than essentially individualistic, holdings would seem to link the structural argument or insight to some distinctly larger goal than structure for structure's sake - typically to a larger goal related to the political system's openness to new ideas and values, its sensitivity to tradition and continuity, and its dedication to evolving norms of decency. One would expect, therefore, that an attempt to reveal a structural logic in Bush v. Gore would tell us about the decision's vision for the polity. If Bush v. Gore was the blueprint for a vessel, toward what shores has it set sail?
But Professor Karlan offers neither blueprint nor destination. She does not suggest that the Court in Bush v. Gore was even attempting, for example, to render the voting process more transparent or accountable or possessed of any other systemic virtue. On the contrary, her claim seems to be that the Court's interference in the vote count was a lunge for power that manifestly obscured more than it clarified. n254 With neither a standing problem to motivate it, nor a systemic aspiration to guide it, the "structural equal protection" that Karlan ascribes to Bush v. Gore seems both directionless and unmoored. Although she may have glimpsed something through a crack in the shell, n255 that was part of the trick: the pea is not where she imagined.
3. Third Shell: Due Process. - Even the Court had difficulty convincing itself that any version of an equal protection analysis by itself mandated rejecting the Florida Supreme Court's recount mechanisms. Instead, the majority turned to what seems to be a due process argument, n256 [*234] pitched somewhat awkwardly and without any real analysis pointing either to the substantive or to the procedural versions of that doctrine.
(a) Procedural? - The idea that the five Justices who comprised the Bush v. Gore majority could countenance a procedural due process argument in favor of that result is particularly farfetched in light of the Rehnquist Court's insistence that no process at all is constitutionally due to an individual unless the state's law first confers some positive entitlement - such as a job held with tenure - that the state itself promised the individual that she would not lose absent some specified cause. n257 The Bush v. Gore majority faulted Florida for failing to provide something that the Rehnquist Court has held countless times is not required by due process: the conversion of an individual interest - here, the voter's undoubted interest in having her vote counted so long as her ballot clearly indicates her intent - into a full-fledged entitlement - here, to a trial-like evaluation of her ballot by an impartial factfinder applying objective criteria. This even though the state has never supplied any objective criteria that could sustain such a conversion. The injury identified in the opinion is the differential treatment afforded "the marks or holes or scratches on an inanimate object, a piece of cardboard or paper ... a thing, not a person." n258 On that basis, if one assumes that state law creates a recounting entitlement, n259 then the process of recounting in the manner provided by state law provides precisely the sort of pre-deprivation "hearing" required by the Court's procedural due process jurisprudence. In any event, it is difficult to see how a counting process that occurs in public view, includes representatives from the interested parties, and guarantees a right of review by a neutral magistrate, could be characterized as fundamentally unfair to individual voters. Admittedly, the state's highest court might have reduced the room for partisan manipulation by the ballot tabulators in the first instance by formulating "objective" criteria of the sort insisted upon by the Court. But whatever criteria the supervisory magistrate might have employed could in principle have reduced [*235] the room for such manipulation at least as much as the criteria that would have been formulated by the state supreme court - probably more, given the inaccuracies that any set of uniform statewide criteria would necessarily introduce. Besides, partisan considerations could just as easily enter into the state supreme court's very selection of the supposedly objective uniform statewide criteria for counting ballots.
Such a selection would face yet another constitutional objection from the five authors of the per curiam opinion if it failed to include those overvotes n260 that were left out of the Florida court's December 8 recount order. That opinion faulted the recount because "three counties were not limited to so-called undervotes but extended [the recount] to all of the ballots," n261 thus treating overvotes in those counties differently from overvotes elsewhere. n262 Perhaps more importantly, a standard that excluded overvotes generally would harm "overvoters" vis-a-vis all "undervoters" in the state. n263 This standard would do the same with respect to "the citizen who marks two candidates, only one of [*236] which is discernible by the machine, [who] will have his vote counted even though it should have been read as an invalid ballot." n264
These putative overvote violations highlight the Court's own errors throughout its per curiam opinion, which focused on saving a tree while burning down the forest. First, while it may be true that three counties' recounts included overvotes, it is also true that thirty-four counties examined overvotes for mistakes during the original machine count and subsequently included, in their initial vote tallies, overvotes that revealed the voter's intent. n265 Despite this fact, the Court tried to have it both ways: If the three counties erred by including their overvotes, then surely the thirty-four counties erred as well. And if the Constitution requires the recount to be invalidated because of selective inclusion, then the initial count must be invalidated on exactly the same ground. But the majority had already elided a true examination of the underlying differences among county voting machines. It was thus no surprise that it focused on an overvote "problem" that could be detected only in the shadow of the larger equal protection problems the Court itself would go on to create.
In fact, the per curiam did not even notice the reflection of the overvote issue in the mirror of its own narrow holding. The Court saw an equal protection violation because overvotes were treated differently from undervotes. But, taking the Court on its own terms, a recount standard that uniformly excludes overvotes should be perfectly constitutional - each overvote is treated the same as every other. n266 Instead, the Court seemed to suggest that overvotes have some positive right to be counted - that, because some "dimpled chads" are considered under positive law, dimpled chads on overvotes must receive that same consideration. But this line of argument leads to an absurd conclusion: voting jurisprudence would no longer be "one person, one vote," or even "one ballot, one vote," but "one dimple, one vote." Each mark would have to be given the same treatment on any ballot on which it appeared - to avoid denying due process to that mark. But it makes no sense to say "a dimple is a dimple is a dimple" when a [*237] dimple next to two punched-through holes may not mean the same thing as a dimple next to two merely dimpled chads.
Because, in any event, the positive law of Florida simply did not permit the counting of overvotes, n267 no Floridian can complain that overvotes were not counted - at least not without raising a substantive right to have one's vote properly counted whenever one's intent is clear. And that standard, after all, is the very one adopted by the Florida Legislature and approved by the Florida Supreme Court but overturned by the Supreme Court's 5-4 per curiam decision on what turned out to be the real "election day."
(b) Substantive? - A more likely explanation is that the Court's demand for "the minimum requirement of nonarbitrary treatment of voters necessary to secure the fundamental right [to vote]" n268 rests on substantive due process. All nine of the current Justices have acceded, in greater or lesser degree, to the practice of infusing substance into what due process of law requires, n269 both by identifying certain spheres of liberty as presumptively beyond government's power to regulate n270 and by delimiting the circumstances in which, or the extent to which, a given exaction may be imposed. n271 But several of the Justices in the bare majority that decided Bush v. Gore are among the members of the Court most reluctant to strike down state action on substantive due process grounds. n272 And even the Bush v. Gore dissenters have routinely expressed "reluctance to expand the concept of substantive due process because guideposts for responsible decisionmaking in this uncharted area are scarce and open-ended," n273 reflecting an unease that has plagued the doctrine from its beginning. n274 Voting rights jurisprudence in particular provides little or no help in anchoring Bush v. Gore to a substantive due process rock, because [*238] state regulations of voting can impinge on liberty in a substantive way only when they restrict the ability of some voters to participate in an election. n275 If the state imposes, for example, a poll tax upon its citizens, in effect treating differently those who can afford and are willing to pay from those who cannot or are not, n276 it denies to indigent voters "the right to elect legislators in a free and unimpaired fashion." n277 But the Florida Supreme Court's "intent of the voter" standard was not a regulation that restricted or burdened citizens seeking access to the polls. On the contrary, it was an attempt to ensure greater voter participation in the election process. n278 Its whole thrust was to prevent [*239] the disenfranchisement urged by Ms. Harris and by then-Texas Governor Bush. But, whether or not one credits that purpose, there is no way to construe the Florida recount as a restriction of any facet of a citizen's substantive liberty.
There is, however, a form of due process that straddles the divide between the substantive and the procedural, a form that draws upon First Amendment analysis to constrain the open-ended power of government arbitrarily to constrain speech or movement. The paradigmatic case is a restraint on speech through a standard so vague or all-encompassing that it permits a government official to reject the speech for any reason or no reason, including its content or even its viewpoint. For example, Lovell v. City of Griffin n279 declared an ordinance invalid [*240] because it "prohibited the distribution of literature of any kind, at any time, at any place, and in any manner without a permit from the city manager." n280 Half a century later, in City of Lakewood v. Plain Dealer Publishing Co., n281 the Court held that a municipality could not burden speech by requiring all organizations that desired to use sidewalk displays to obtain prior approval for their designs from the city architect. This procedure was objectionable in its application to newspapers, in part, because it afforded city officials broad discretion to reject conduct closely related to protected speech on the basis of its content, through the application of vague standards such as "reasonableness." The lack of statutory guidelines left the Court with no principled means of distinguishing a government official's legitimate denial of a permit on content-neutral grounds from the "illegitimate use of censorial power." n282 Similarly, an anti-loitering statute is void when it "authorizes or even encourages arbitrary and discriminatory enforcement." n283
The basic thing to notice about this "hybrid" due process right is that it has been limited primarily to the regulation of speech, expressive association, and freedom of movement. In these contexts, officials must be limited to a narrowly circumscribed and objectively defined [*241] set of grounds on which they may restrict the individual's liberty. But outside these three contexts, the norm has been considerable tolerance of official discretion. n284 Indeed, Justice White, in a dissenting opinion joined by Justices Stevens and O'Connor, once noted that the only cases in which the Court had struck down as facially unconstitutional a law delegating overly broad licensing discretion to local officials involved activities the locality could not prohibit altogether, like leafleting on public sidewalks. n285 And even Justice Brennan's majority opinion in that same case, joined by Justice Scalia, among others, agreed that laws conferring on local officials discretion over the issuance of building permits or other licenses unrelated to ongoing expression carry little danger of censorship and accordingly are not subject to facial review. n286
But, as the Court expressly stated in Bush v. Gore, "the individual citizen has no federal constitutional right to vote for electors for the President of the United States unless and until the state legislature chooses a statewide election as the means to implement its power to appoint members of the electoral college." n287 If the right to be free from discretionary decisions by officials is limited to activities the state must permit, then that right does not extend to the counting and recounting of votes for presidential electors. Even if the right is not so limited, it certainly does not extend beyond the realm of constitutionally protected activities that might be subjected to well-hidden censorship of political, religious, or other views. But there is no demonstrable danger that canvassing officials would deliberately (but unprovably) refuse, during a recount, to give ballots their clearly intended meaning because those officials preferred other candidates. And it is especially unlikely that voters "will [feel] pressured to conform their speech" - even assuming voting could be equated with speech n288 - "to the [canvassing official's] unreviewable preference" n289 so as to maximize the prospects that their ballots, if they were ever recounted, would be counted properly. The ex ante prospect of a recount is too low to make such calculated behavior plausible, and in [*242] any event it would be absurd to vote for a candidate you oppose in order to reduce the risk that a hostile canvassing official will misrecord your ballot! That would be cutting off your nose to spite your face. Subjectivity and discretion, rather than being constitutionally condemned (and compelled to give way to "neutral criteria to insure that the ... decision is not based on the content or viewpoint of the speech being considered"), n290 are thus constitutionally permissible here - especially when held in check by public observation and judicial review.
The Court had certainly never before extended the requirement of narrowly drawn, determinate standards to anything resembling the process of counting (or recounting) ballots. n291 From districting to ballot counting, the government processes involved in voting are open to public scrutiny, and it is that scrutiny that polices their integrity. Accordingly, the evils to be avoided in vote counting are opacity and corruption. So long as transparency is maintained, due process requires only a standard reasonably calculated to establish the intent of the voter.
Three further points are worth making. First, because state involvement is necessary to the honest and efficient conduct of elections, a much greater degree of state discretion in controlling the voting process is tolerated. n292 Therefore, substantive due process concerns cut little ice. For example, in Burdick v. Takushi, n293 the Court refused to strike down laws requiring all candidates to participate in primary elections and preventing voters from casting "write-in" ballots to nominate candidates. The Court held that primaries provide sufficient access to the electoral system for all candidates and that the burden imposed on voters whose preferred candidates failed to participate in the primary was insufficient to violate voters' rights of free association. n294 Nor was there any burden on the voters' right to speak. A voter does not express any particularized view about a given candidate when she casts a ballot, other than her endorsement of that candidate, or slate of candidates, for public office. n295 For example, the voter need not identify with much - or any - of the candidate's platform, may vote for the candidate only because she does not like the others, or [*243] may even vote for the candidate by mistake. n296 Voting, especially when each ballot is cast in secret, is not so much a matter of engaging in positive acts of speech, as it is a matter of taking part in the process of "winnowing out and finally rejecting all but the chosen candidates," ... not [a matter of] giving vent to "short-range political goals, pique, or personal quarrels,'" n297 but of participating in a collective political enterprise. The voting process provides, at best, a structured mechanism into which individuals can insert tokens of their preferences for political candidates. These are then aggregated to yield a result. It may be that the voter can make whatever political statement she wants by scribbling on or spoiling the ballot, or demonstrating outside the polling booth. But by its very nature, a properly cast ballot does not make any "statement" apart from its role in the selection of candidates to represent the voters in government. Here, the Florida Supreme Court's standard is better understood as an attempt to ensure the fair and honest counting of ballots than as an unconstitutional limitation on the range of political opinions which voters may freely express or from among which voters may freely choose. n298 As such it should not have been subject to the strict scrutiny that the Court implicitly employed in Bush v. Gore when it demanded the creation of uniform nondiscretionary, mechanically applicable statewide standards. n299
Second, even if we were to treat each recounted ballot as a tiny burst of speech, the Florida standard would survive strict scrutiny because the ballot counter's only delegated authority is to interpret marks, holes, or indentations on small pieces of cardboard as one of a limited set of statements: vote for Bush, Gore, Buchanan, Nader, or nobody. No authority is delegated to the vote counter to approve or disapprove the voter's choice. The risk that this power to construe might be perverted into a power to censor is no more significant than the risk that the power to deny a building permit by invoking neutral zoning criteria might be used to deny permission to an unpopular newspaper. n300 In the Court's only prior encounter (of which I am aware) with regulations of objective characteristics of speech-bearing documents - there, the permissible size and color of photographic reproductions of U.S. currency - the Court upheld the regulations, [*244] which appear to have been enforced by an eyeball test rather than by a spectrometer or some other mechanical device or objective formula, and did so under a reasonableness standard, in sharp contrast to the strict scrutiny under which the Court struck down the same statute's criminal prohibition of using photographic reproductions of U.S. currency except for "educational, historical, or newsworthy purposes" - an exception the Court held impermissibly content-based. n301
Even on a matter completely removed from the First Amendment and its cousins (such as the vital decision whether or not to continue life-sustaining care), Justice O'Connor, who seemed particularly impatient during the oral argument in Bush v. Gore with the very idea that any voter who did not properly follow the explicit and objective instructions might have a case for consideration of her vote, n302 argued in her concurring opinion in Cruzan v. Missouri Department of Health n303 that "states which decline to consider any evidence other than ... explicit oral or written instructions regarding [individuals'] intent to refuse medical treatment should they become incompetent ... may frequently fail to honor a patient's intent" and, in so doing, violate "a duty [that] may well be constitutionally required to protect the patient's liberty interest in refusing medical treatment." n304
That we do not invariably use little pieces of cardboard for expressing our wishes should we become incapable of doing so later on seems irrelevant to the question whether the Due Process Clause requires the use of mechanical formulae or procedures, whenever possible, to eliminate the risk that a government decisionmaker will allow his or her own personal preferences regarding what an individual should intend to influence how that decisionmaker understands, or at least records, that individual's expression of what he or she does intend with respect to a matter of fundamental right. If anything, the use of rigid formulae, because they are bound to produce the wrong results on many occasions, even if in a random manner, might well be deemed constitutionally suspect whenever a fundamental right is involved - something the Court has indeed held with respect to the determination of fetal viability in Colautti v. Franklin, n305 with respect to a pregnant woman's capacity to perform her job in Cleveland Board of [*245] Education v. La Fleur, n306 with respect to a man's fitness to be a father after the death of his child's mother in Stanley v. Illinois, n307 and with respect to a number of rules regarding residency and travel. n308
Third, to the degree that voting is thought to be sui generis in this regard, it is noteworthy that a requirement of the sort the Court posits in its per curiam opinion would with one stroke invalidate the results of most of the elections held throughout American history, including the presidential election of 2000, inasmuch as most of the ballots in all of those elections were tallied by hand. Machines for tallying votes are, after all, a recent innovation. n309
To be sure, the Court did insist that its holding was limited to the supposedly peculiar circumstance of a statewide recount under the supervision of a single judicial body that possesses the theoretical capacity to demand uniformity but fails to do so. n310 But whatever one might think of the supposed distinction between such a recount and the initial counting process for purposes of the Court's other arguments - and, as I make clear elsewhere, I think very little of it n311 - one context in which it cannot possibly make any sense is the supposed right not to have a vote counter's preferences influence how that vote counter treats an individual's ballot.
If there were such a right - and if it entailed an ancillary instrumental right that the counting entity use a mechanical formula that determines how to interpret dimples, chads, and all the rest - then that right would not depend on what the vote counter knew about the results from other precincts, or whom NBC was projecting as the winner [*246] nationwide. n312 For the right in question would require more than a guarantee that whoever counts your vote be ignorant of other election data. It would have to include, centrally, a guarantee that whoever counts your vote not take his or her own preferences as to the outcome into account in deciding what you meant to convey by some ambiguous mark or sign on your ballot. Given the impossibility of reducing to zero the chances of this occurring and of having all violations of this standard detected and corrected by observers or a reviewing court, the supposed right would necessarily require a mechanical method - one whose errors would certainly be random and nonpartisan - applied to each ballot in order to translate it into a vote. Any such right could not depend on whether the translation occurs on the morning of election day, just before the polls close, hours after the polls have closed in a precinct that counts simultaneously rather than on a rolling basis, or in a statewide recount under the supervision of a single judicial magistrate.
It follows that, if there were such a right, it would have been violated in all elections in which manual methods were used under an "intent of the voter" standard in the initial count or in a recount, a standard that must surely be widespread even today. n313 Although it is impossible to say how many election results would have to be tossed out (or at least recorded as invalid even if left undisturbed) as a consequence, those as close as that between George W. Bush and Al Gore on November 7, 2000 surely would be. It follows that no right of the sort the Court posited has ever existed in this country or could be recognized without condemning a significant body of legal and political results.
Crucially, even the most determinate rule does not in itself prevent biased counters from influencing decisions. The Florida vote-counting process responded to just such a threat - that counters might see one thing and tally another thing altogether - by having all interested parties send observers and by authorizing a neutral magistrate to [*247] break any ties. Scrutiny is directed where it counts - to the practice of counting votes, rather than to the verbal or numerical formula by which ballots are supposedly transformed into votes to be counted.
The pea is therefore assuredly not under the due process shell either. We have returned to where we started: with a group of disillusioned Florida voters unconvinced by the Court's equal protection rationale in any of its guises. And we're left with a badge that rightly proclaims: "EQUAL PROTECTION, MY ASS!"
4. The Irresistible Allure of Appearances. - That there are no serious doctrinal contenders for the rationale underlying the Court's holding does not mean that it was inexplicable. The ruling's privileging of the interests of atomized individuals at the expense of political and historical reality is symptomatic of the current Court's peculiar form of "constitutional denial." The Court routinely fixes its gaze upon the superficial appearance of propriety or impropriety in the way individuals interact with government without ever genuinely coming to grips with the difficult issues lurking beneath the surface. All must look prim and proper, whatever the underlying realities.
It is by now a trite analysis of the Rehnquist Court's race-related jurisprudence to point out that the majority often avoids considering a historical pattern of discrimination against a group, focusing instead on the discrete injury suffered by an individual. n314 Within this framework, the injuries suffered by whites and by members of racial minorities appear to be of essentially the same kind, and our "color-blind Constitution" requires that they be treated equally. n315 Thus, Bush v. Gore certainly does not stand alone in the pantheon of decisions rendered [*248] by this Court with one eye focused firmly on the appearance of things and the other eye just as firmly shut to the underlying reality. n316
In McCleskey v. Kemp, n317 the Court found that a pattern of racial discrimination that rendered a black defendant charged with killing a white victim nineteen percent more likely to be sentenced to death than a white defendant charged with killing a black victim n318 did not demonstrate "an unacceptable risk of racial prejudice influencing capital sentencing decisions." n319 The chance that race had been illegally used as a proxy for reasoned decision in these cases was seen as nothing more than a statistical "discrepancy," the sort of "apparent disparity ... [that is] an inevitable part of our criminal justice system." n320 The Court seemed to find the brutal consequences of that disparity - the fact that "McCleskey's claim that racial bias has impermissibly tainted the capital sentencing decision ... easily could be extended to apply [throughout the process of capital sentencing]" - too much to contemplate. n321 The Court thus appeared to be paralyzed by the prospect of too much justice n322 - the idea that there was "no limiting principle to the type of challenge brought." n323 Instead of confronting the racism pervading the system of capital sentencing in Georgia, the Court saw racism's very pervasiveness as a reason to ignore it.
The Court reacted differently when, during the sentencing phase of a capital trial, the prosecutor introduced evidence that the defendant, a white male, was a member of the Aryan Brotherhood, a white prison gang organized to ""respond' to gangs of racial minorities." n324 In an opinion by Chief Justice Rehnquist, the Court held that, if "the murder [*249] victim was white," membership in a white racist prison gang was not evidence of bad character admissible at the sentencing phase, n325 and vacated the death sentence that the jury had imposed. n326 This focus on the race of the victim in the particular case missed the point of the evidence: that the defendant's membership in a prison gang that had pledged to harm minorities and that had in fact done so in the past made it more likely that he had also engaged in such acts and thus that he had precisely the propensities and posed precisely the dangers that Georgia deemed relevant in deciding which capital defendants should be sentenced to die. n327 The jury found this evidence sufficient to impose a death sentence, but the Court could not see past appearances to recognize the validity of that inference: anyone looking just at the race of this white victim would see no evidence of racial bias by this white defendant.
To be sure, appearances sometimes do matter. Sometimes, how government action looks to a reasonably well-informed but disinterested observer is relevant to - even dispositive of - constitutionality. The type of injury at stake when appearance is itself central has been described as "expressive harm": injury "that results from the ideas or attitudes expressed through a governmental action, rather than from the more tangible or material consequences the action brings about." n328 Expressive harms typically result from government action open to public view, as in the endorsement or rejection of a particular group or its ideas based on characteristics normally subject to strict scrutiny. n329 Much of the scholarly work on expressive harms has focused on Shaw and its progeny - cases in which districts have been apportioned based on racial criteria. n330
[*250] The strand of Establishment Clause doctrine focusing on a state's endorsement of religion illustrates particularly well the relation between the message seemingly expressed by a government action and its constitutionality. n331 In that context, the Court has held, rightly I believe, that the appearance of neutrality is vitally important and that a state should not "foster[] or encourage[]," but must instead "take[] pains to disassociate itself from the private speech" of religious groups. n332 Thus, the display of a creche inside a courthouse is distinguishable from a display composed of a Christmas tree, a plainly secular sign from the mayor, and a menorah outside the building: the message one display conveys to an informed but disinterested observer arguably differs greatly from that conveyed by the other. The question in each case is whether ordinary citizens "will perceive the combined display of the tree, the sign, and the menorah as an "endorsement'" or "disapproval ... of their individual religious choices,'" judging the matter "from the perspective of a "reasonable non-adherent.'" n333
Even if Justice O'Connor did not hold the pen that authored Bush v. Gore, n334 it is noteworthy that for over a decade, she has been the Court's most consistent champion of expressive harm as a fulcrum for constitutional analysis. In the areas of voting rights, minority set-asides, and sex discrimination, she has used an "expressive" analysis to attack as unconstitutional various acts "stigmatizing" groups along constitutionally suspect lines. Best known, perhaps, is her decision in Shaw, forcefully stating the view that "reapportionment is one area in which appearances do matter. A reapportionment plan that includes [*251] in one district individuals who belong to the same race, but who are otherwise widely separated by geographical and political boundaries, and who may have little in common with one another but the color of their skin, bears an uncomfortable resemblance to political apartheid." n335 And in Adarand v. Pena, n336 Justice O'Connor, writing for the majority, struck down a government set-aside "because [the] perception - especially when fostered by the Congress of the United States - [that special preferences should be granted to some groups on the basis of race] can only exacerbate rather than reduce racial prejudice." n337
She has advanced a similar line of reasoning in the gender context. Thus, in Mississippi University for Women v. Hogan, n338 Justice O'Connor's majority opinion rejected as illegitimate the differential treatment of male and female nurses, reasoning that such treatment projected the "archaic and stereotypic notions" that women "are presumed to suffer from an inherent handicap or to be innately inferior." n339 More recently, Justice O'Connor has extended the reasoning of Hogan into the realm of immigration law. She dissented from a decision to treat differently for purposes of deportation someone whose mother was an American citizen at the time of the individual's birth and someone whose father was an American citizen when the individual was born. n340 In that case, she argued that, by regarding mothers and fathers differently, Congress impermissibly relied upon sex-based stereotypes that expressed approval of outdated and illegitimate conceptions of gender roles. n341
Like these cases drawn from realms as diverse as the separation of church and state and gender classification, Bush v. Gore lays great emphasis on appearances - both of individual ballots and of the vote-counting process as a whole. n342 Unlike the other cases in the Court's [*252] oeuvre in which appearances have mattered, however, Bush v. Gore was a case in which the appearances the Court emphasized did not single out a class of "people by changing their relationship to the State," n343 whether by virtue of their race, their gender, their religion, their political views, or any other characteristic in terms of which the government might "send[ ] a message to [some] that they are outsiders, not full members of the political community, and an accompanying message to [others] that they are insiders, favored members of the political community." n344 Election 2000 presented no danger of any such exclusion or marginalization. On the contrary, the Florida Legislature and state supreme court, recognizing that the relationship between openness and participation is part of both the message and the challenge of democracy, were attempting to meet that challenge by taking the process of counting votes out of the inner workings of the tabulating machinery and placing it in human hands, before human eyes, for all the world to see. Indeed, part of the vote count's legitimacy was expressed through the public's opportunity to watch it and to have proxies, in the form of party members and a supervising magistrate, there to represent the public and partisan interests in a fair count. "People in an open society do not demand infallibility from their institutions, but it is difficult for them to accept what they are prohibited from observing. [Opening government activities to public scrutiny creates] an opportunity both for understanding the system in general and its workings in a particular case." n345
What, then, was the matter? No proponent of the Court's equal protection theory has, of yet, articulated how any appearance of inequality, irregularity, or instability projected by those intrepid ballot readers caused anyone to feel marginalized, diminished, or otherwise reduced to less than full membership and participation in the polity. Bush v. Gore, one might think, has Shaw's fingerprints all over it, sharing Shaw's appearance-driven modus operandi and focusing, as Shaw did, on the "strange" look and language of the voting process. Shaw's misshapen districts n346 have been replaced in Bush v. Gore by those omnipresent hanging, detached, and dimpled chads. n347 But, whatever one might think of Shaw as an exemplar of expressive harm to an excluded or marginalized group - were African-Americans insulted by their inclusion in the district? were whites insulted by their exclusion? - at least it reflected a comprehensible concern with government action that "reinforces the perception that members of the [*253] same racial group ... think alike" and "bears an uncomfortable resemblance to political apartheid." n348
Nothing of the sort warrants a parallel concern about appearances in Bush v. Gore. There, the Court's equal protection theory evinced at best a concern for appearances at the most abstract level of equality imaginable: "to sustain the confidence that all citizens must have in the outcome of elections." n349 Missing from the Bush Court's opinion - and from the writings of its most articulate defenders n350 - is any explanation of the mandate from which the Court draws its authority to wield the Fourteenth Amendment not as a shield for individuals who might be oppressed or made to feel like outsiders, but as a sword to cut short a functioning democratic process. The Court, sitting astride the body politic, pronounced an ongoing, publicly transparent, judicially supervised election not "good enough for government work" - a bit too messy and, well, political to keep up some Court-prescribed level of popular confidence in the results of the election. Will we hear next of the vital role of "miracle, mystery, and authority" n351 in sustaining the people's faith in their elected leaders?
It would be one thing if the Court had intervened to quash an appearance of bias in a recount for white voters, for Protestant voters, or for voters who agreed to restrain their expression in other ways. It is something altogether different when the Court intervenes, as it did in Bush v. Gore, to protect some overarching appearance of partisan neutrality and a decorum that by its very nature flourishes in shadow and is upset by daylight. Such intervention grafts an entirely new branch onto the thriving tree that grew into the Fourteenth Amendment from the blood-soaked soil at Gettysburg. What brooding omnipresence anointed these nine worthies to rewrite Lincoln's immortal words - that government of the people, for the people, but by the Court, shall not perish from the Earth?
In Bush v. Gore, the Court mistook measures reasonably designed to enhance accuracy and equality for a system that reeked of bias. In removing the vote from the people to prevent that supposed bias from operating, the Court removed the very bulwark against both actual bias and its appearance - public scrutiny - that was necessary to legitimize the Court's own impact on the process. When the Court permits appearance to become the dominant value even though what is at stake is not just the appearance of justice, but its very essence, it leaves justice undone.
[*254] It is not an option to trot out various excuses for the Court - that it did not have enough time to think through its decision fully or that its reasoning was truncated. The Court's decision, however prospectively limited, extends backwards into its civil rights and First Amendment jurisprudence to evince a deep-seated concern with how things look, rather than how things are; with the appearance of "propriety" - instead of with truth or justice or all those outmoded ideas - as the measure of "constitutionality." n352
B. Speaking Visions to Theory - The Four Dimensions of the Seen and Unseen
The succession of television images through which the nation came to know the case did more than simply bring us the story of Bush v. Gore; it almost certainly shaped the U.S. Supreme Court's understanding of the stakes involved and its ultimate holding that the Florida Supreme Court's December 8 order - which mandated a statewide recount of undervotes n353 but gave no specific rule-like guidelines for their evaluation, merely instructing election officials to discern the intent of the voter if clear - violated the Equal Protection Clause. Take, for instance, the televised close-ups that became the enduring legacy of the entire dispute: the close-ups of election officials holding partly perforated ballots up to the light, squinting to see what dents or holes or hanging chads they might detect, and quarreling over how to tally those ballots as possible votes in the presidential race. For those exposed to this dizzying, almost intoxicating, barrage of images accompanied by punchy scrolling text lines, it quickly became axiomatic that ballots identical in appearance might end up being counted differently in different counties or in the same county at different times. In time-pressured, breathlessly conducted litigation, with unheard-of deadlines crowding what would normally be six weeks of briefing into a weekend and what would normally be months of judicial reflection into a matter of hours or at most days, those pictures were worth not thousands, but tens of thousands, of words.
1. Invisible History. - One crucial aspect the TV screen could not show was the unique history lying behind each of the similar-looking ballots that the election canvassers were inspecting visually on camera. Given those distinct histories, uniform statewide counting rules would necessarily introduce their own distinct inequalities. In one precinct, for instance, more of the punch-card voting machines might be outdated [*255] or in poor repair; or the chads might have been less recently cleared out of the balloting machines; or more of the voters might be elderly or frail; or voting instructions, when required by law to have been printed in Spanish as well as English, might have been printed in English only. n354 For an election canvasser in one precinct to count a mark or a perforation as a vote under a more forgiving standard than one applied by a canvasser in another precinct, where an identical-looking ballot is likely to have had a different history, could in actuality honor, not violate, the principle of "one person, one vote." When one sees only the ballot being held in the air and scrutinized with squinted eyes, it may be difficult to remember that the principle is "one person, one vote," not "one ballot, one vote." One must construct in the mind's eye a vision of the person who cast that ballot and the story connecting that person to the machine that registered no vote for president when he or she encountered that machine at a particular time and place.
Florida's highest court, from its vantage point closer to the action, presumably recognized the futility and counterproductivity of any effort to mechanize, standardize, and ultimately dehumanize the process of assessing the voter's probable intention from a ballot's appearance. Imagine, for instance, rigorously enforcing a rule that provides that a punch-card ballot counts only when the chad has been detached on at least two corners. Suppose that, in one county, the ballots were printed on such thick cardboard that no chads were detached on more than a single corner, and most were merely indented. Or suppose that, in another county, hundreds of ballots were found with no detachments but with the name "Bush" or "Gore" clearly circled, each ballot obviously violating the instructions but clearly revealing the voter's intent. If you reply that the uniform standards approach would, of course, make exceptions for such cases, then you've given up on the quest for objective criteria articulable in advance. For the only way to decide which cases merit exceptional treatment is to demote the uniform rules to the status of mere guidelines, all to be applied with a dollop of subjective judgment and a pile of plain old common sense.
But essays on the virtues of flexible standards don't sell copy, and rational and painstaking deliberations don't condense well into the scrolling text lines and soundbites that define the modern media. It was no surprise, then, that this made-for-television production found no way to communicate the impossibility of formulating uniform rules for accurately translating into votes the infinite variety of marks that voters, including those who wrote in names, might have left on the [*256] ballots in pencil, ink, or with a stylus. n355 Besides, if uniform and accurate statewide rules could be framed, surely the single judge assigned by the December 8 order to review the county-by-county recounts would be as well positioned as the state's highest court to frame them. Yet only Justice Stevens, dissenting, thought that "the fact that a single impartial magistrate will ultimately adjudicate all objections arising from the recount process" could alleviate the concerns raised when different counties use similar voting systems but "differing substandards for determining voter intent." n356 Although that task might have been performed imperfectly, any claim that it would be was plainly not ripe for adjudication. In any event, straining all of these multifaceted background facts through a uniformly metered sieve would necessarily have yielded an unequal stream of results, not a stream that flowed evenly in accord with the one-person, one-vote formula. But the ever-present spectacle of election officials holding ballots in the air to divine each voter's intent eclipsed the potentially harmonizing image of subsequent review by a state magistrate, while the illusion of a technical "fix" in some uniform set of rules that the state's high court was faulted for not having promulgated orbited brightly overhead like a mechanical moon. The Justices who suggested that machine-like algorithms could workably replace human judgment in deciphering machine-rejected ballots were no doubt tuned in to breakthroughs in artificial intelligence and computerized pattern recognition unknown to the rest of us.
2. Comedy of Errors. - Another dimension relegated to obscurity by the TV screen - one not exactly hidden from the TV viewing audience yet seemingly lost on the Court - was the significance of the fact that election officials can err in at least two diametrically opposed directions: they can wrongly fail to count a vote that someone intended to cast, or they can wrongly attribute a vote to someone who either intended to cast no vote at all or, worse still, intended to cast a vote but for an opposing candidate. Machines can do that too, of course, as they did to the Palm Beach citizens who intended to vote for Gore and unwittingly became Buchanan "supporters." Similarly, an error during a manual recount can turn a "none of the above" voter into a phantom supporter of one of the leading candidates ("false inclusion"), or can [*257] turn a supporter of either of the candidates into an imaginary supporter of the other ("false reversal"), or can turn someone who intended to vote for one of those two into a "didn't cast any vote for President" statistic ("false exclusion"). This range of possible errors during a manual recount (false inclusion, false reversal, or false exclusion) makes it impossible to treat differing degrees of leniency in the recount standard used in different places that have the same kinds of vote tabulation equipment as though they corresponded to differing degrees of accuracy in discovering previously uncounted votes. n357 The Court implied that the use of a more stringent recount standard in the more Republican-leaning Palm Beach County violated the "one person, one vote" principle by diluting the votes of its residents relative to the votes of those who resided in the more Democratic-leaning Broward County, which used a "more forgiving standard." n358 Citing the fact that Broward County "uncovered almost three times as many new votes [as in Palm Beach], a result markedly disproportionate to the difference in population between the counties," n359 the Court suggested that any recount technique with a lower per capita yield of "new votes" in a given county would somehow mean a higher disenfranchisement rate for the voters of that county. n360 In truth, no such equation is possible.
The Court in effect allowed the ephemeral multiplicity of types of errors to be obscured by the illusory concreteness of geographic comparisons. Once we eliminate that mirage, it becomes clear that no uniform set of statewide counting standards could begin to account for all the differences in the design and thickness of ballots; in the form and maintenance of tabulation equipment; and in the age, political leanings, and other demographic characteristics of distinct areas yielding errors of either false exclusion or false inclusion (including its most extreme variant, false reversal). No uniform set of standards can in the end minimize the aggregate number of errors or come any closer to approximating what the Court deemed the applicable ideal of "one person, one vote."
But the Court was prepared to sweep whole categories of contested ballots under the rug of a "uniform statewide standard" rather than let counties count their ballots according to guidelines tailored to each county's unique conditions. Perhaps false exclusions, false inclusions, and false reversals would not have canceled one another out. Still, two facts remain indisputable: First, it remains unclear which counties' [*258] standards, if any, would have diluted the voting power of their citizens. Second, any such dilution would have occurred not at the hands of a statewide official deliberately trying to discriminate against a particular segment of the population, but at the hands of the county's own canvassing board. This board would have had no incentive to "dilute" the votes of its own voters. Without such ex ante intent, there is no discrimination. Thus, the Court should have been hard pressed to declare a recount constitutionally infirm when the most certain outcome of voiding it would be to ensure the invalidation of votes that were "legal" under Florida law.
3. The Untold Story of the Underlying Count. - The Court also failed to grapple with the multiple defects in the underlying vote count that its decision left in place. This brings us to the third dimension: the comparative analysis so conspicuously missing from the Court's opinion. Six of the seven Justices who suggested that there were equal protection problems in the recount process seemed to assume that they could validly put to one side, and relegate to another day and another lawsuit, the multitude of errors and inequities that made a mockery of the underlying vote count leading to the certification of Bush as the winner by 537 votes out of over six million cast. n361 Some have argued [*259] that, because the irreducible margin of error exceeded that minuscule margin of victory, simple prudence dictated action by some grown-up invoking whatever excuse might be cobbled together to halt the match without further ado. But any such view must assume that the underlying vote count was entitled to be treated as at least comparatively authoritative, despite all the flaws in the process generating that vote count - including ample opportunities for state and local officials to tailor the processing of voters and the handling of both live and absentee ballots to the latest election updates from East Coast broadcasters, keeping in mind how those officials' adoption of one practice rather than another might shape the outcome - and despite the probability that adjusting the underlying count with the results of the recount, with its supposed flaws or with the state's being given time to correct them, would at least yield a more reliable determination of who won.
To offer just a few telling examples of why one might question the underlying vote: Florida law unambiguously mandated at least one initial recount in all sixty-seven counties because the vote totals in the initial count came within one half of one percent of each other. n362 Yet eighteen of Florida's counties conducted no such recount. n363 And there is no doubt that differences among counties' voting systems, many of which were understood to be positively correlated with socioeconomic status, ethnic background, and political affiliation, not only rendered the initial vote count less reliable than Florida law required, but also led to a politically and racially skewed distribution of error rates. Thus, for example, far more ballots were rejected in the twenty-four punch-card counties (3.9%) than in the twenty-four optical-scan counties with second-chance capability (0.6%); in the fifteen optical-scan counties without second-chance capability, the rejection rate was highest of all (5.7%). n364 Minority voters were roughly ten times more likely not to have their votes correctly counted in this election than were nonminority voters, n365 and that's to say nothing of the larger if more difficult to document phenomenon of obstacles that prevented minority voters from even obtaining a ballot that anyone could count. [*260] In some of the forty-one optical-scan counties, ballots were counted as valid in the initial counting process even when voters simply checked or circled a candidate's name; n366 other counties excluded all ballots that were not machine-readable. n367 Moreover, eleven counties used the infamous "butterfly" ballot (Palm Beach, for example) or some other confusing form of "wraparound" ballot that no amount of recounting could translate into a vote mirroring the voters' actual intent. n368
Yet, despite the obvious errors and inequities in the underlying count, only Justice Ginsburg's dissent posed the logically crucial comparative question: what in the Court's opinion (or in the dissenting opinion of Justice Souter n369) demonstrated "that the recount adopted by the Florida court, flawed as it was, would have yielded a result any less fair or precise than the certification that preceded that recount?" n370 Or, I would add, any less fair or precise than a recount under whatever uniform set of specific rules for particular ballot configurations the Florida Supreme Court might have fashioned on remand? The inescapable answer is: nothing. n371
It is not as though the problems with the underlying vote count and the certification embodying it might somehow be swept under a rug emblazoned with the motto, "Another problem for another day." Because a presidential election cannot be rerun after its flaws have been corrected, n372 to strike down the recount is to uphold the count itself: the two are joined at the hip and cannot be severed. Talk of the "fundamental right of each voter" cannot be permitted to obscure the big picture: the aggregation of votes into a single result is what counts. Yet the Court obscured that picture by subtly transposing the wide-angle panorama before it - a montage in which one candidate sought to secure electoral victory by halting the statewide recount of legal votes - into a close-up shot of a beleaguered voter at the mercy of officials free to apply an "intent of the voter" standard as they saw fit. At the end of the day, the only reason given in the Court's per curiam opinion for failing to approach the problem comparatively, and thus focusing solely on the arrangement of the deck chairs rather than on [*261] the plight of the Titanic itself, is that "the problem of equal protection in election processes generally presents many complexities" - complexities supposedly, but not genuinely, absent when one focuses on "the minimum procedures necessary to protect the fundamental right of each voter in the special instance of a statewide recount under the authority of a single state judicial officer." n373
Nor is it enough to argue that the virtues of decentralized decisionmaking might overcome the inequities of the initial certification, as the majority evidently assumed and as Justice Souter seemed to suggest. n374 First, even if the interest in equality of opportunity with respect to matters like public schooling may properly be sacrificed on the altar of local autonomy (as it was in San Antonio Independent School District v. Rodriguez), n375 the reason is only that the Court has yet to treat even equality of educational opportunity - despite its proximity to voting - on par with equality of the franchise, the right that the Court said the Florida recount sacrificed. Second, any interest in localism that the Court deems cognizable with respect to differences in underlying voting mechanisms, ballot designs, or automatic recount practices should be no less cognizable with respect to recount procedures under the December 8 order. There is simply no difference for equal protection purposes, from the perspective of decentralized decisionmaking, between the count and the recount. If there were, it would be a nice question on which side of the line to locate the automatic recount mandated by Florida law but actually undertaken in just forty-nine of Florida's counties in this election. n376 Third, whatever [*262] weight any Justice may have given to the distinctively legislative character of the judgment to decentralize decisionmaking with respect to the underlying vote, it is clear that the terms of the Florida Supreme Court's recount order derived, as every Justice but the Rehnquist three presumed, from the legislation enacted to discharge the state's duty under Article II, Section 1, Clause 2.
Finally, some have suggested that insisting on strictly scrutinizing inequalities in the vote caused by variations in local methods and machines, along with the inequalities on which seven Justices focused in Bush v. Gore, requires us also to scrutinize strictly other inequalities among counties that give some groups - principally those who have greater wealth and belong to no disadvantaged racial or ethnic minority - more political power than others. n377 I very much doubt that defeating the distinction between the initial vote and the recount would open any such Pandora's box. The most glaring arbitrariness in the Court's equal protection analysis was its willingness to subject intercounty variations in the manual ballot recount to strict scrutiny while disregarding altogether intercounty variations in the machines and methods for counting ballots in the first instance. Not only did the Court disregard these other variations, but it also declared them beyond the reach of the precedent it was establishing. Moreover, this arbitrariness is different in kind, and not simply in degree, from the arbitrariness that some might say inheres in subjecting differences in the treatment of ballots (whether in the initial count or in various recounts) to strict scrutiny, while applying less searching scrutiny to differences among localities in public funding of education or in other aspects of government policy that undoubtedly contribute to race-or class-based differences in meaningful opportunity for political participation.
[*263] To be sure, there might seem to be no basic distinction between the fact that relying exclusively on local property taxes to fund public education indirectly results in residents of property-poor districts being less able to make full use of the franchise than are the residents of property-rich districts, n378 and the fact that voting mechanisms and procedures in poorer districts or in largely minority districts are more onerous, or less likely to tally every voter's intent accurately, than are voting mechanisms and procedures in wealthier, largely white districts. But ordinary intuition, and the lines the Court has drawn for decades in its equal protection jurisprudence, converge in this instance to support a distinction between inequalities in the voting system itself and inequalities in the infinite web of government policies and practices that might be shown to impinge directly or indirectly on that system and on the ability of various individuals or groups to exploit fully that system's potential. This distinction, though perhaps imperfect, seems defensible in this context. And it would surely be indefensible to permit the virtual impossibility of judicially drying up the ocean of injustices that somehow directly affect the right to exercise one's franchise, and the great difficulty of guaranteeing full equality of voting opportunity, to excuse the Court's decision to halt a manual recount of ballots because of the supposed inequalities it contains without even considering how doing so could yield a net reduction in voting equality given the clearly more severe inequalities in the way the votes were initially tallied by machine. The injustices with which the Court seemed so concerned, including the supposed due process plight of individual ballots, existed when the ship set sail on November 7 and were exacerbated, not reduced, when the Court stopped the recount dead in the water on December 9.
4. Ballots as Castaways. - After its unjustifiably narrow focus on the constitutional defects in the December 8 court-ordered statewide recount, and despite all its talk about the "minimum procedures" that must be followed in such "special instances" to preserve the "fundamental rights" of individual voters, the Court, in devising a remedy for the alleged equal protection and due process violations, decided that the Titanic's deck chairs might as well go down with the ship, too - something that becomes even clearer when we examine the fourth dimension, which the television cameras did not capture: the invisible spectacle of thousands upon thousands of ballots being jettisoned into oblivion by the Court's order that the recount stop.
The Court's only justification for ending the recount, rather than at least allowing the Florida Supreme Court to try to fashion a remedy for the supposed defects in its December 8 order, was the ostensible [*264] decision of that court that state law, embodied in the election code, precluded counting past December 12. Far from second-guessing the state's highest court on the meaning of state law, as many of the Court's critics mistakenly accuse it of doing, the Court just guessed at what the state court would say, if asked, about whether Florida law permitted recounts in a presidential election to continue past the December 12 safe harbor or instead required recounts to stop at midnight on December 12, whatever the cost in potentially decisive legal votes that would remain uncounted. The Court claimed that it had to defer to the Florida Supreme Court's supposed finding that the Florida Legislature, to ensure that Florida could "participate fully in the federal election process," wished to avail itself of the safe harbor offered by 3 U.S.C. 5. This federal statutory provision indicated that Congress would accept without challenge the presidential electors from any state that by December 12 had fully and finally resolved, in accord with the "judicial or other methods or procedures" in place on election day, any election-related "controversy or contest." n379 And the Court treated that imagined state judicial finding as a mandate to end the recount by December 12, come what may. n380 To many whose main mantra had been [*265] that the U.S. Supreme Court should defer to the Florida Supreme Court on all matters of Florida law, that twisting of the knife should have brought to mind the famous maxim: "Be careful what you wish for. You just might get it!"
Chief Justice Rehnquist's concurring opinion, joined by Justices Scalia and Thomas, did not adopt the per curiam's deferential posture toward the state's highest court. The Chief Justice instead argued that, in conferring power specifically on the state legislature, the Constitution's mechanism (Article II, Section 1, Clause 2) for directing each state's choice of presidential electors limited the degree to which the Court could properly defer to the state judiciary's reading of the state's election code. n381 But the concurring opinion's "independent" analysis of Florida's election statutes led to the same bottom line: given the benefits of resolving all election contests in time to fit within the safe harbor, the Florida Legislature, in empowering state courts to grant "appropriate" relief, "must have meant relief that would have become final by" December 12. n382 This in turn meant that the Florida Supreme Court had violated Article II by ordering a recount on December 8 that could not possibly have been completed by December 12, leading the Chief Justice to conclude that the November 26 certification of Bush as the winner by 537 votes had to stand as the final result. n383
That dog won't hunt. You can read the Florida statutes - which deal with presidential and gubernatorial elections in the same set of provisions - backwards and forwards without finding even the slightest clue that the state legislature ever decided that all recounts in a contested presidential election must stop by December 12 or, for that matter, at any time before the electors meet to give their votes on December 18, n384 or even before Congress starts to count the votes on [*266] January 6. n385 And there was certainly nothing to suggest that the Florida Legislature had ever considered, much less resolved, the question of what to do when confronted with a tradeoff between the benefits of the safe harbor and those of counting all the legal votes. There is an air of particular unreality in the suggestion that the Florida Legislature that enacted the 1999 election contest provision at issue in Bush v. Gore had thereby legislated a mandatory preference for the benefits of the safe harbor over the imperative of counting every legal vote cast in the election. For the members of that body were, with very few exceptions, the same as those of the Florida Legislature in 2000 - and it was that legislature, as everyone knew, that stood poised on December 11 to name its own electoral slate committed to Bush after the December 12 deadline had passed, n386 apparently with the Court's blessing. Moreover, the Florida court's own alleged treatment of December 12 as a drop-dead deadline, on which the per curiam opinion relied, was at best highly ambiguous. n387
Clearly Florida's legislature wanted Florida's electors counted in the national tally, all other things being equal. But not even the Supreme Court in its per curiam ruling purported to attribute to the [*267] Florida Supreme Court a specific decision to end all presidential election contests in time to come within the December 12 safe harbor. Nor could the Florida Supreme Court have possibly equated the general wish to have the state's twenty-five electoral votes counted with a specific December 12 deadline. Even concluding all election contests by the December 12 safe harbor date would not have guaranteed that Congress would honor the promise made by the 1887 Congress that passed the Safe Harbor Act. n388 As Justice Ginsburg's dissent noted without rebuttal from the Court, the 1887 Electoral Count Act contained later deadlines and alternative ways of bullet-proofing an electoral slate for the congressional count. n389 The one thing anyone could say for sure was that it wasn't the Florida Supreme Court's December 8 recount order but rather the U.S. Supreme Court's December 9 stay of that recount order that deprived Florida of a shot at the benefits of the congressional safe harbor. For - and this is surely the crowning irony - it was that very stay that guaranteed that Florida could not possibly produce by December 12 any final resolution, using the rules and procedures in place on November 7, of the electoral contest that the stay had interrupted. The "final" resolution that the December 12 per curiam decision rushed to generate appeared to come just in the nick of time, with two full hours to go before the safe harbor shut its gates to new state vessels filled with electoral votes. But the Court's less-than-deliberate speed was to no avail: the resolution it yielded was produced not by Florida using that state's November 7 rules but by the U.S. Supreme Court using its (newly minted) version of the Equal Protection Clause! For all that it mattered, the Court could have splurged and spent an extra three or four days, quadrupling the amount of time devoted between the argument and the decision (and, who knows, maybe written a cogent opinion - perhaps even reached a different decision n390) without any loss whatsoever vis-a-vis the safe harbor, whose benefits the Court itself had rendered unattainable on December 9. n391
[*268] Worse still, let's pretend with the Court that halting the recount on the afternoon of December 9 and pronouncing it dead late on December 12 allowed Florida's electors, as previously certified on November 26, to sail right into that safe harbor. Then, reading the laws of Florida (or, for that matter, of the United States) to mandate this purchase of electoral security at the price of ditching into the sea thousands of ballots, many of which clearly expressed the intent of the voters who cast them, would by the Court's own criteria have been flatly unconstitutional. Justice Stevens, dissenting, made the point tellingly. As he observed, the Court had acknowledged that, under Florida law, "all ballots that reveal the intent of the voter constitute valid votes." n392 Thus, under the Court's own reasoning, the Fourteenth Amendment precludes any legislative or judicial determination valuing the state's interest in finality above the right of voters to have all valid ballots counted. As the Supreme Court had told the Florida high court: "The press of time does not diminish the constitutional concern." n393 And it would beg the question to say that the rights of all the voters, including those whose ballots would never be counted, would be safeguarded by shielding from challenge the electors certified on November 26. The whole question was whether counting the remaining ballots would have shown that the wrong electors had been certified! Thus, whether the objection is based on Article II or on equal protection, the decision to toss out all the remaining legal votes as of 10 p.m. on December 12, 2000 was utterly indefensible.
C. Good for This Day and Train Only
1. Is There a Future for Bush v. Gore? - Predictably, a cottage industry of glowing possibilities is already under construction, using as its bricks and mortar the Rehnquist Court's seeming embrace in Bush v. Gore of voting rights precedents dating from the Warren era that some feared might never be dusted off and taken down from the [*269] shelves again. n394 Ever hopeful, some civil rights groups have already launched ambitious new lawsuits building on equal protection theories they see as implicit in Bush v. Gore. n395 I would be the last person to discourage such efforts, which at worst would shine welcome light into the hall of mirrors where Bush v. Gore and its reflections might too comfortably remain obscured. Nor would I venture to predict that this singular decision will become and forever remain a "derelict on the waters of the law." n396 For the day might yet come when a pearl is drawn from that oyster. But those are waters in which I wouldn't hold my breath - not just because the political pendulum still has many swings to go through before the wish to put Bush v. Gore to some "progressive" or "liberal" use overtakes the federal bench, but because, if my analysis is correct, there simply is no "there" there: as our constitutional shell game revealed, Bush v. Gore belongs to no constitutional tradition and grows out of no constitutional doctrine, principle, or clause. Lacking roots, it is a plant anyone would be hard pressed to grow.
Nor is this circumstance an outgrowth simply of the Bush Court's insistence on circumscribing the decision as narrowly as it did - a matter to which I turn next. For even if the Court had invited all and sundry to extend its recounting holding to, say, intercounty variations in voting machines and methods, the sources of doctrine for any such foray would necessarily be decisions like Reynolds n397 and Harper n398 and perhaps Boddie v. Connecticut. n399 Bush v. Gore might provide the ornamentation but could not truly provide the analytic framework or the constitutional groundbreaking for a critique of those variations. The mechanistic, Sesame-Street-like catechism on which the Bush syllogism rests ("If both of those ballots look just the same, what must [*270] their votes be like?") is simply too shallow, its gruel of superficiality too thin, to furnish a full jurisprudential meal.
2. The Value of a "One Time Only Equal Protection Offer". - That a hypothetical Bush v. Gore stripped of the "design defect" of an artificially circumscribed holding might be of hardly greater value as precedent than the real Bush v. Gore cannot, however, exonerate the Court from the charge that it sliced the salami (or the baloney) just a bit too thin. Any court, after all, has an obligation to render principled decisions, not rulings whose edges are as carved and compromised as the least coherently crafted legislation. Nor is it enough to say in its defense that almost any judicial ruling involves line drawing, and that although one can disagree over where the cut should be made, the Court's solution was as plausible as any other. For the Court decreed an election outcome that transmuted the underlying count into a presidency and, in the process of so doing, created divisions - between the count and recount; between December 12 and December 18; between December 18 and January 6 - where none in fact existed in reality. Each stage was one in a continuous process that did not end (and could not end in a constitutional sense) until Congress had tabulated the votes of the electors. The Supreme Court, in drawing its illusory count/recount line, n400 simply ignored the partisan influences and inequalities that shape the process at every stage n401 by virtue of the choice of ballot type and format, the selection of vote tabulation systems, and the treatment of voters. n402 The Court evidently forgot or chose to overlook the way in which each stage operates as a single act in a performance that does not end until the last curtain calls are announced by Congress. Thus, the "line" along which Bush v. Gore would have us divide the electoral world turns out not to be a line at all, but a mirage. n403
3. Unprincipled or Merely Indefensible? - If the Court is successful in limiting its equal protection rationale to "the present circumstances," n404 it will have effectively precluded Bush v. Gore from having any precedential value. Elections that fall within the margin of error calling for vote recounts are extremely rare, and advances in vote-counting technology, spurred in large part by this past election, will hopefully make hanging, dimpled, and pregnant chads things of the past. Because the prospect of an ex post statewide recount under similar [*271] circumstances may never again arise, the effect of Bush v. Gore on legal doctrine could prove to be a nullity, despite the best efforts of voting rights advocates to leverage it into a brave new world of voter equality.
Many see the Court's attempt to limit the case to whatever "the present circumstances" might be as profoundly illegitimate. n405 These critics argue that the Court was in essence trying to free itself from the discipline of stare decisis, which forces a court either to eat its own words in future cases or else give good reasons for spitting them out. By shirking that discipline, the Bush v. Gore Court, they continue, abandoned the claim to legitimacy that can come only from feeling, and announcing that one feels, bound tomorrow by what one says today. n406 Indeed, whenever an Article III court renders a decision, these commentators argue, that decision must have precedential effect. It is no good for an Article III court, particularly the Supreme Court, whose function is to lay down rules for lower courts, to employ reasoning that resembles a one-way, nonrefundable railroad ticket, good for this day and this destination only. n407 Some ground this type of objection in Article III, n408 whereas others are more inclined to ground it in prudential considerations. n409
No matter what the underlying rationale, these objections to Bush v. Gore amount to a claim that the decision was so unprincipled and result driven as to be an exercise in something other than true adjudication. By ushering in a new President and then dispensing with the reasoning by which it reached that result, the Court effectively had its [*272] cake and ate it, too: it installed its chosen President in office, and it won't have to bother with pesky new civil rights lawsuits premised on Bush v. Gore's equal protection reasoning. In geometric terms, we might represent court decisions as points in space. As cases are decided, more points appear in that space; these points may be connected with lines or surfaces, which we call precedent. Precedent allows us to guess, with varying degrees of accuracy, depending upon the fit of the line or surface, where the next dot may fall in space, and to argue about where it should fall. n410 Critics argue that by limiting its decision so strictly, the Court in essence declared Bush v. Gore an isolated data point - and because an infinite number of lines can pass through a single point, the decision is unprincipled.
While there is undoubtedly some truth to these objections, they overstate the point to a degree that makes them quite vulnerable. First, while the term "judicial power" in Article III is not wholly devoid of content, it certainly is not so full of meaning as to prescribe a certain level of precedential value to which all decisions must rise. Second, although it is appealing to say that judges and Justices should be bound by the decisions they render and the rules they make, such a principle makes a much better guideline than a rule. What does it mean to require that a decision have precedential effect? In very few situations can an opinion truly be said either to have or not to have precedential value. The question of what precedential value to accord an opinion is nearly always more nuanced, a matter of degree. n411 The precedential value of a decision is more closely likened to a continuum than to a binary variable. There seems to be no principled way to draw a line on that continuum n412 and thus no principled way to criticize [*273] an opinion a priori solely on the ground that it was too narrow. n413 There are times when great problems inhere in a system, and the only remedy is a systemic approach; at other times, great problems arise in a single case and are best remedied by simply resolving that case. Both are legitimate functions for a court of last resort.
In the end, drawing a line between the principled and the unprincipled or between the precedential and the nonprecedential is like drawing a line in water. At the extreme, a decision limited to a narrow set of facts or circumstances only by a "principle" that is so internally inconsistent or incoherent as to "imply" anything and therefore nothing n414 is indistinguishable from a decision literally limited to the case at hand and hence from a wholly unprincipled adjudication - an illegitimate exercise of power masquerading as judicial. Any decision about how to characterize Bush v. Gore - whether to deem it a holding circumscribed by a line that gropes toward a principle but falls apart upon inspection, or a holding that utterly lacks principled justification and is an adjudication in name only - might best await what the Court makes of Bush v. Gore in future cases and whether it treats that ruling, either openly or by its silence, as a "designer decision" concocted solely for the occasion. And it can be something more only if the Court does two things:
First, it must revisit the illusory line Bush v. Gore purported to draw between inequalities in the recount process and inequalities in the underlying count.
Second, the Court must develop a framework for assessing these underlying inequalities that looks for its substance not to the appearance-driven, glittering surface that is Bush v. Gore, but to the more nurturing earth beneath - to Reynolds, Harper, and their legitimate heirs.
[*274]
D. Dissonance
And so, as Mozart's Emperor Joseph II was fond of saying, "there it is!" n415 We are left with an account of the Court's action so simple that it is almost an embarrassment: its holding, it seems, best comports not with the traditional canons of equal protection doctrine, n416 nor with its fancier modern counterparts, n417 nor with any of the variations of due process, n418 but rather with a kind of sound-and-light show. A barrage of images - the all-too-fallible human beings squinting into the lights, the partisan loyalists storming into recount facilities, and the stacks upon stacks of ballots left to be counted by fatigued and imperfect counters - struck for the Bush v. Gore majority a dissonant constitutional chord. n419 The Court reacted to the dissonance without waiting for its inevitable resolution - a resolution bound to take place in the "Finale: prestissimo" movement of The Election 2000 Symphony - by responding in its all too familiar and simple way. Upset with an appearance it found distasteful, the majority took responsibility and "stepped up to the plate." n420 That vision of responsibility in the face of supposed chaos has its basis not in any defensible conception of the Court's role in a democracy - whether an attempt at neutral explication of text and structure; or the representation-reinforcing, minority-protecting, perfection of democracy; n421 or the protection of an evolving concept of human dignity n422 - but rather in a much more straightforward, albeit disheartening and condescending, assumption that the [*275] Court simply knows best and that what's best is what the Court thinks looks best.
V. The Mystery of the Court's "Unsought Responsibility"
Are we expected to overlook or forgive the shallowness of the Court's argument just because of the insufficiency of the record and the time pressures under which the Court was forced to discharge its "unsought responsibility," n423 as the per curiam opinion put it? "Unsought responsibility"? No way. The Court had the temerity to suggest that, simply because the "contending parties invoked the [judicial] process," it had no choice but "to resolve the federal and constitutional issues [it] had been forced to confront." n424 Really? The cases are too numerous to cite in which the Court, extolling the virtues of abstention or avoidance, has let pass from its lips a cup laced with far less poison than this. For the Court to lecture the nation about "the vital limits on judicial authority" and to tell it that "none are more conscious of [those] limits ... than are the members of this Court," n425 in a case in which the Court had reached out to grant discretionary review - without an explanation as to why it, rather than Congress, is the proper decisionmaker - is insult enough. Expecting the nation to endure that lecture from a Court that felt obliged to insist that "none stand more in admiration" than it does "of the Constitution's design to leave the selection of the President to the people, through their legislatures, and to the political sphere," n426 goes over the top.
What, precisely, was the Court's justification - even assuming that its ruling on the merits was defensible - for first interrupting, and then bringing to a close, the national conversation that was Bush vs. Gore, for making it into Bush v. Gore and then calling the game before the matter could reach Congress? Florida Chief Justice Wells, dissenting from his court's 4-3 recount order of December 8, quoted a mathematician's view that the "margin of error in this election is far greater than the margin of victory, no matter who wins." n427 Perhaps so. With every recount, the state would emerge with just another conclusion, and not necessarily one more closely approximating the truth. For Wells, given the futility of identifying an unambiguous winner, the game wasn't worth the candle. Worse, "prolonging [the] judicial process [*276] in this counting contest [would have] propelled this country ... into an unprecedented and unnecessary constitutional crisis" that would "do substantial damage to our country, our state, and to this Court as an institution." n428 That's not a bad approximation of the most sophisticated among the published defenses (that of Judge Posner) of what the Supreme Court did - if not of what it said in agreeing to review Bush v. Gore - by issuing its stay and then ruling for Bush in a manner that brought the election to a close on December 12.
A. The Forgotten "Rule of Law": The Political Question Doctrine
The Supreme Court was less forthright than Chief Justice Wells but appears to have had much the same idea in mind. Exactly what "crisis" it feared on the road ahead, however, remains something of a mystery. I part from the cynics who are confident that the "crisis" the Court was determined to avert was the presidency of Al Gore. Had the Court cared only about blocking a Gore victory, it could readily have calculated, based on what was known at the time, that simply letting the December 8 recount continue would probably have accomplished that goal. If the recount had yielded a victory, however narrow, for Bush, that would have ended the matter. Had it instead yielded what Florida's highest court was prepared to certify as a Gore victory, there was every indication that the Florida Legislature stood ready to certify a slate of Bush electors notwithstanding any such judicial decision. Indeed, the Court's December 12 opinion strongly suggested that the state legislature could do so at any time. n429 I think the Court should have been more circumspect about the state legislature's authority to take such ex post facto action once the votes had been cast and electors, albeit of disputed identity, had been chosen. The relevant congressional legislation empowers a state legislature to direct the choice of a new set of electors only when the state "has held an election for the purpose of choosing electors, and has failed to make a choice on the day prescribed by law," n430 not when it has made a choice but the choice has yet to be ascertained. Be that as it may, only the Court would have been in a position to stop the state legislature from construing its constitutional mandate more broadly, and the Court must have been aware that it could have simply declined to enjoin the state legislature when it certified Bush the winner.
Furthermore, even setting the state legislature aside, who would have been in a position to stop Florida's Secretary of State, a Republican who had co-chaired the Bush campaign, from submitting to Congress [*277] the slate of Bush-pledged electors she had certified on November 26? The Florida Supreme Court? Hardly. If that court had sought to interfere with the submission of a slate of electors, either by the Secretary of State or by the state legislature, the U.S. Supreme Court would have had a far clearer legal basis for intervention than it could possibly have conjured up on December 9. For in that event, the Court would have been acting to prevent a state's nullification of federal supremacy by direct obstruction of Congress's ability, under the Twelfth Amendment, to determine which set of Florida electoral votes to count. n431 Supreme Court intervention at that point would have rested squarely on the Supremacy Clause of Article VI, the glue joining the several states together into a single Union.
The odds must surely have seemed great, had the Supreme Court opted to remain on the sidelines apart from acting as necessary to ensure that no one stood in the way of Congress's final resolution of the matter, that Bush would still emerge victorious. The 1887 Electoral Count Act contained a labyrinthine procedure setting forth how the Congress that convened on January 6 was to resolve disputes over election results in contested states and how it was to break any remaining ties. It is true that even that procedure, untested in the 114 years during which it has been in place, was shadowed by constitutional doubt over the power of one Congress to bind its successors in such matters. n432 But, even so, the relative power of the Republican and Democratic Parties in the Congress made it likely that, in the end, Bush would have emerged the winner.
There is a powerful case indeed for the Court playing no role other than to protect Congress's decisionmaking function - that is, for treating the matter as a political question textually committed to Congress under the Twelfth Amendment, rather than a legal question properly resolved by a court. n433 The requisite textual commitment n434 [*278] to a political branch could hardly be clearer. n435 Yet the majority evidently thought it unnecessary to justify the Court's seizing the last word on a matter that involved the identity of the next President, for it didn't so much as mention the political question doctrine.
Not materially different from Article II in this respect, the Twelfth Amendment specifies that the President of the Senate shall "in the presence of the Senate and House of Representatives" open all the certificates containing the votes of the putative electors and that "the votes shall then be counted." n436 When the normal operation of the electoral college failed to yield an undisputed winner in the Hayes-Tilden presidential election of 1876 and in the congressional consideration that ensued in 1877, Congress ended up creating a special electoral commission to resolve the matter. The commission's fifteen members included five sitting Supreme Court Justices and ten members of Congress, and its decisions could be vetoed by the two Houses of Congress acting concurrently. Today, of course, the service on such a body by members of Congress would be understood to violate the separation of powers as construed by Buckley v. Valeo, n437 and the reservation of a veto power in Congress acting by anything less than full legislation presented to the President for signature or veto would be understood to violate the nonparliamentary structure of our government. n438 But although constitutional doctrine has been unkind to the precise form of dispute-resolving institution Congress chose to create to untie the Hayes-Tilden knot, neither constitutional doctrine nor constitutional history has cast any doubt upon the paramount place Congress occupied in creating that institution to deal with the conflict.
There were then and remain today significant ambiguities as to the Twelfth Amendment's allocation of authority for counting the electoral votes. The Amendment's text says only one thing directly relevant to [*279] the matter: "The President of the Senate shall, in the presence of the Senate and House of Representatives, open all the [electoral] certificates and the votes shall then be counted." n439 The Framers should have listened to the time-honored injunction to avoid the passive voice. "Shall then be counted" - by whom? And what discretion does the vote-counting individual or entity have to reject one elector or slate of electors and accept another in the elector's or slate's stead? Is the President of the Senate vested with whatever authority the power to "count" entails? Is any such authority reposed instead in one or another House, or in the two Houses acting concurrently, or in the two Houses acting as a single organ even though not precisely as the Congress of the United States? These are complex questions on which the text, structure, and history of the Amendment are, it seems fair to say, less than decisive. But that same text, structure, and history are entirely decisive in establishing that power to resolve electoral disputes - to decide which electors were duly selected to represent any given state in the manner that state's legislature directed in accord with Article II, Section 1, Clause 2 - is not entrusted to the Chief Justice of the United States, to the Supreme Court of the United States, or to any other officer or part of the judicial branch of the United States. The House Committee reporting the bill whose ultimate embodiment was the Electoral Count Act of 1887, of which the famous safe harbor provision was a part, had no difficulty concluding that "the power to determine [contests over competing electors or electoral slates] rests with the two Houses, and there is no other constitutional tribunal." n440 That conclusion tracks the plain language of the Twelfth Amendment, and of Article II before it, and sounds like a "textually demonstrable constitutional commitment of the issue to a coordinate political department" n441 if ever there was one.
How remarkable was it that neither the Court's per curiam opinion nor the Chief Justice's concurrence so much as mentioned the political question issue, much less attempted to justify its assertion of authority in the face of the seemingly applicable political question doctrine? It's hardly the sort of thing a Supreme Court Justice simply forgets about. And even if it were, the briefs called the attention of the Justices to the problem. n442 Besides, Justices Breyer and Souter held their colleagues' [*280] feet to the fire on the point, the former quoting directly from the House Report quoted above. n443 It's tough to avoid the inference that the five Justices in the Bush v. Gore majority were essentially speechless on the matter. Had they spoken up in defense of their seizure of jurisdiction in the face of a textual commitment so seemingly plain, I suppose the Justices would have argued that nothing about the political question doctrine has ever been so cut and dried; that the doctrine leaves the Court with considerable leeway in deciding when to invoke it as a decision avoidance device; and that the great urgency of resolution by a fully functioning and credible branch of the national government counseled against decision avoidance here. Having argued earlier in this Comment that there was no such great urgency, I am obviously unlikely to be much moved by this imagined counterthrust. But even if I reached the opposite conclusion on the urgency issue - something that quite a number of eminent lawyers and thinkers have, after all, done - I would be most uncomfortable with the whole line of argument. For we are not talking here about some discretionary zone within which the Court may properly exercise, or decline to exercise, the passive virtues n444 of abstention from decision, as the Court so often does in denying certiorari even though a matter is within its jurisdiction and may meet the technical requirements of a conflict in the lower courts or some other pressing need for Supreme Court intervention. Nor are we talking merely about some sort of judge-made "doctrine" serving to fill in the vast open spaces of the Constitution. This space is fairly well closed.
Justice Thomas testified before a subcommittee of the House several months after the decision in Bush v. Gore, proclaiming that "if there was a way ... to have avoided getting involved in that difficult decision ... I would have done it." n445 Justice Thomas is far too bright to have been brainwashed by the per curiam opinion's pretense that the Court had no choice in the matter. Except that, in a sense, it did have no choice in the matter: the only lawful choice, not because of any theory of passive virtues or because the counsel of prudence so dictated, but rather because the Constitution so commanded the Court, was not to inject itself into the dispute.
[*281] If the constitutional command were not enough, the counsel of prudence would not have been far behind. Justice Breyer's dissent ably canvassed the reasons for believing that involving the Justices in the choice of presidential electors is simply "out of the question," n446 as no less an authority than James Madison put it long ago. Drawing both on that conclusion and on the consequences for the Court of having the Justices play a role even pursuant to congressional mandate in the Hayes-Tilden affair, Justice Breyer praised the "decision by both the Constitution's Framers and the 1886 Congress to minimize this Court's role in resolving close federal presidential elections." n447 "However awkward or difficult it may be for Congress to resolve difficult electoral disputes, Congress, being a political body, expresses the people's will far more accurately than does an unelected Court. And the people's will is what elections are about." n448 Moreover, awkwardness and difficulty are not to be found entirely on Capitol Hill.
One should not underestimate how awkward and difficult it must have been for the Court to play a decisive role in the selection of the President. However false, irrelevant, or both all accusations of partisan motive may be, n449 the Court cannot shut its eyes to the inevitable presence of such accusations - and, among those disinclined to voice those accusations, worrisome doubts. Precisely that sort of consideration has figured in prior decisions by the Court to treat a matter as political rather than justiciable - even where the Constitution's text, augmented by a statutory scheme already in place for implementing the responsibility that text assigns to Congress, is considerably more opaque than is the case with the Twelfth Amendment. Thus, the Constitution's provision that the Senate is to have the "sole" power to "try" all impeachments n450 must be stretched substantially to support the Court's conclusion that what constitutes a "trial" is a nonjusticiably political question, committed solely to the Senate, foreclosing jurisdiction in the federal courts of a claim by an impeached and convicted federal district court judge that the proceedings conducted by the Senate did not amount to a trial at all. n451 When the Court read the constitutional language giving the Senate the exclusive jurisdiction to conduct trials of impeached federal officers as a preclusion of the power of judicial review, it had to be influenced - it said it was, n452 and how [*282] could it not have been? - by the appearance of impropriety attaching to any picture of the Court deciding whether an impeached federal judge, a judge of the judicial system at whose apex the Court sits, had been properly convicted and removed from office by the Senate. There seemed no way to read the Constitution to preclude judicial review in such cases while permitting it when nonjudicial officers have been impeached by the House and convicted by the Senate, so the solution was to preclude judicial review across the board, treating as political what would ordinarily have been a straightforward legal question.
Compared with the question whether a given proceeding amounted to a "trial" - a matter surely within the comparative legal expertise of judges and Justices - the question whether an electoral slate chosen by a certain recount method treats the voters fairly and represents the people of a given state, could hardly look more political. And if the specter of sitting in judgment on a brother or sister jurist from the lower federal bench reeks of impropriety or the appearance thereof, what is one to say of five Justices deciding who might name their successors? The problem, I repeat, is not confined to situations in which one can say that Justices in that predicament have in fact permitted themselves to be swayed in their legal judgments by their political wishes. I will make plain shortly that I claim no such thing. The problem, rather, is one of appearances. For a Court so sharply focused on how each government action appears to insiders and outsiders alike, n453 it is ironic that, in this case, it took so little care to minimize the appearance of its own impropriety. n454
Altogether different would be a federal constitutional challenge, brought in advance of the presidential election, to a state's scheme for choosing electors, alleging that the design of the scheme offends Article II, Section 1, Clause 2, or the Equal Protection Clause, or any other constitutional provision or principle. Nothing in the Twelfth Amendment suggests that resolution of that kind of dispute is committed unreviewably to Congress. On the contrary, the procedure sketched in the Twelfth Amendment, and later implemented by the Electoral Count Act of 1887, does not give Congress the authority to jump into the fray in anticipation of the next election, before there are any electoral "certificates" to open or any "votes" to count, in order to weigh in on the question of a state electoral scheme's constitutional validity. Nicely illustrating the point that the "political question" doctrine might not fare too well in the "truth in advertising" department, this scenario underscores the fact that the same abstract legal question might be [*283] embedded in two differently structured controversies, in one of which the question is justiciable and in the other of which it is not. n455
Thus it is certainly imaginable that at least some of the questions tendered for decision in Bush v. Gore, such as the consistency with the Fourteenth Amendment of the recount scheme that the Florida Supreme Court interpreted the state's election code to provide, would have been justiciable if posed in advance as challenges to the regime under which Florida proposed to conduct an impending presidential election. n456 Indeed, the Supreme Court's seminal precedent involving Article II, Section 1, Clause 2, the case of McPherson v. Blacker, n457 briskly rejected a claim by the State of Michigan that the constitutionality of using districts to choose electors was a nonjusticiable political question n458 even when raised in a purely pre-election posture rather than in the heat of battle. Michigan had argued in conclusory terms that "all questions connected with the election of a presidential elector are political in their nature" and that this case was "not of judicial cognizance" because "its decision would be subject to review by political officers," including "finally, the Congress," which remained free to count electors chosen from districts even if the Court proclaimed that mode of choice unconstitutional, and free not to count electors so chosen even if the Court gave that mode of choice its constitutional blessing. n459 Just as that argument was rejected in 1892, so it continues to lack force in our era - a point illustrated by the fact that a similar argument was not taken seriously in United States v. Nixon. n460 In that case President Nixon had argued that, because the [*284] Court could not force him to turn over the damning Watergate Tapes inasmuch as he could simply fire one set of executive officials after another until he found an Attorney General compliant enough to rewrite the regulations that granted the special prosecutor independence and authority to contest presidential claims of executive privilege, it followed that any judicial decision was subject to political revision by the executive branch and that the question was therefore nonjusticiable. n461 The Court easily rejected the argument. n462
The upshot is that pre-election constitutional challenges to state systems for conducting presidential elections need not be political rather than justiciable in character - and a good thing, too, since no other federal tribunal would be available at that stage to consider the constitutional argument - while challenges to a state's actions in the course of a particular presidential election, challenges that reach the Court during the election or so close to the election that it appears no decision other than one stepping on Congress's Twelfth Amendment toes would be possible, should be regarded as political rather than justiciable.
It would be interesting to know what light the Bush v. Gore majority would have shed on this analysis had it deigned to address the matter. But, as is old news by now, the Court simply took for granted its authority to weigh in and essentially to decide the presidential election, undeterred by the direct challenge in the dissents of Justices Souter and Breyer to its authority to do so.
Judge Posner rationalizes the Court's resolution of this decidedly political question by painting Bush v. Gore "as a pragmatic solution to a looming national crisis." n463 Conjuring a highly unlikely doomsday scenario that culminates in the appointment of Treasury Secretary Lawrence Summers as Acting President - some crisis! - Posner suggests that the Supreme Court's intervention did nothing less than stabilize the global economy and maintain world peace. n464 Indeed, "the single most arresting feature of the litigation to a pragmatist is the [*285] worst-case scenario that it avoided." n465 After emphasizing the need "to balance the good effects of steady adherence to the "rule of law' virtues ... against the bad consequences of failing to innovate when faced with disputes to which the canonical texts and precedents are not well adapted," n466 Judge Posner's lopsided characterization of Bush v. Gore as necessary to avert national, and perhaps even global, calamity is surprising. Yet this "crisis-averted" characterization is a necessary element of his defense of Bush v. Gore, for by his own admission, "the Constitution had to be stretched to provide a remedy for Bush." n467
Recognizing the infirmity of the Court's equal protection analysis, n468 Posner would have preferred that the Court stretch the constitutional fabric at Article II rather than at the Fourteenth Amendment. n469 Judge Posner does not justify reversal under Article II solely on the ground that the Florida Supreme Court misinterpreted Florida election law, n470 and my disagreement with Posner and the Bush v. Gore majority is not limited to a polite parting of ways regarding niceties of the Florida election code. Rather, I vigorously contest Judge Posner's, and the Court's, assumption that stretching the constitutional fabric was necessary to protect the nation itself from being torn apart. Judge Posner analogizes Bush v. Gore to the infamous decision in Korematsu v. United States, n471 in which the Supreme Court upheld the internment of Japanese-Americans during World War II. The cases are analytically similar, according to Posner, because they illustrate how "powerful norms of legal justice ... can bend to practical exigencies." n472 The comparison is apt: Judge Posner's dire depiction of the fragile state of American democracy on December 12, 2000, is as dubious [*286] as the Korematsu Court's endorsement of racially driven internment as a military necessity during World War II. n473
If I am correct that, as the Court's majority saw the prospects on December 9, Bush seemed likely to emerge as the next President regardless of what the Court did, then its decision to intervene rather than to respect the teachings of the political question doctrine, and its decision to stop the counting permanently, seem more than adequately explained - but not justified - by the Justices' obvious dismay at the processes that lay ahead, both on the ground and in Congress. The Court's dramatic intervention also reflected an openly expressed desire to leave the anticipated Bush succession unmarred by what Justice Scalia described, in an opinion accompanying the Court's decision to issue a stay, as a "cloud upon what [Bush] claims to be the legitimacy of his election." n474
What cloud?
Presumably, it was the cloud that would form from the very knowledge of the mounting number of possible votes for Gore, perhaps even a number high enough to overturn the razor-thin edge held by Bush as of December 9. But that concern seems irreconcilable with the First Amendment, inasmuch as the Court could have granted review while denying the stay, leaving itself the option of later setting the record straight by announcing that the recounted votes had not been tallied in a lawful manner. Freedom of speech and of the press quite plainly prevent government from suppressing information that lies at the political core of free expression on the theory that the minds of adult citizens would be irreversibly polluted by learning how the nonlegal votes stacked up because the reasons for disqualifying that tally could never be adequately explained. In the regime of the First Amendment, the remedy of refuting information the government deems erroneous or misleading is conclusively presumed to be sufficient.
If the Court had stayed out of the controversy, leaving ultimate resolution to Congress, it would have sacrificed the ability to clear away, to the limited degree it thought possible, whatever shadow might be cast by what it deemed illegal votes. The Court evidently trusted neither Florida's highest court nor Congress nor the people to [*287] disregard improperly counted votes even as Congress named Bush the next President. n475 In essence, the difference between remaining on the sidelines and leaping into the fray (even without granting a stay) was giving up the chance to "edit" the way the political pundits and the history books would describe the margin of the narrowest of Bush victories. Because that is too gossamer a difference to explain something as momentous as the Court's intervention in this case, it follows that jumping in while letting the recount proceed was never a real option for the five Justices in the majority. What they responded to was the very process, from the visual inspection and manual recounting underway in Florida to the anticipated posturing and dealmaking in Congress, that would otherwise be likely to run its course before Bush could be inaugurated as president. The Court's action accordingly emerges as a decision to short-circuit the elaborate set of political processes that otherwise lay ahead en route to that highly probable, if not altogether inevitable, destination.
B. A Herculean Pretension n476
What seemed so distressing about that set of processes? To these five Justices, images of ballots being counted across Florida in front of a nationwide television audience - images that, for many others, conjured memories of the voting rights struggles of decades past n477 - must have evoked thoughts of chaos, partisan manipulation, and mob rule. And the prospect of Congress's serving as the tribunal of last resort must - to Justices who, as we shall shortly see, have anything but respect for that body - have conjured images of corrupt logrolling and irresponsible posturing. Anything so visibly threatening to stability, good order, the established hierarchy, and the appearance of decorum necessarily qualified as a "crisis" that fell within the Court's inherent jurisdiction to halt. Never mind the less open and visible [*288] partisan maneuvering that had led to the radical maldistribution of reliable voting machines and methods among Florida's sixty-seven counties, and never mind such other less dramatic sources of inequality and political favoritism as the subjective and variable standard for deciding which absentee ballots to accept. Those machinations, evidently less disturbing to the Court because submerged below the surface of public visibility, preserved political stability of a kind rather than threatening it. These five Justices were, after all, the very ones who had perceived the Court as the sole institution capable of ensuring political stability in Timmons v. Twin Cities Area New Party, n478 the case in which I argued unsuccessfully for a First Amendment right of multiple political parties to nominate the same candidate - a practice known as "fusion" - that the Court treated as threatening to the stability-producing hegemony of the two major parties.
The key to understanding this Court is to recognize its staggering confidence in its own ability to define and prioritize values of constitutional magnitude and decide which measures are needed to realize those values. To judge from what this Court does, not what it says, high on that list of values is the preservation of a stable order and of an appearance of regularity. Low on that list is an energized, politicized, unruly electorate struggling to find its way toward concrete outcomes in such forms as the election of a president.
The Court's self-confidence in matters constitutional is matched only by its disdain for the meaningful participation of other actors in constitutional debate. Many commentators (including those as otherwise divergent as Michael McConnell n479 and I n480) read the Constitution to give Congress substantial latitude in defining constitutional terms for purposes of exercising power under Section 5 of the Fourteenth Amendment. This perspective proceeds from a recognition that Congress is not invariably confined by the same institutional and federalism-based constraints properly inhibiting the Court from broadening the definition of rights secured by Section 1 of the Fourteenth Amendment. Despite such arguments, the Court has been rigidly doctrinaire in dismissing any pluralistic or dialogic approach to constitutional interpretation and enforcement. n481 It has struck down in rapid succession key provisions of the Religious Freedom Restoration Act, n482 [*289] the Patent Remedy Act, n483 the Trademark Remedy Clarification Act, n484 the Age Discrimination in Employment Act, n485 and the Americans with Disabilities Act n486 - all because it concluded that Congress had gone beyond what it regarded as the correct understanding of Fourteenth Amendment rights or that Congress, having hewed to the Court's understanding, had taken more far-reaching steps than the Court deemed necessary.
A similar hubris underpinned Dickerson v. United States, n487 in which the Court gave the back of its hand to Congress's attempt to legislate around the holding of Miranda v. Arizona. n488 What was striking about the Court's decision was not that a majority, led by Chief Justice Rehnquist, ultimately invoked principles of stare decisis to conclude that Miranda's warnings, as watered down by intervening decisions, n489 had become such a familiar part of the "national culture" that their constitutional validity should not at this late date be reexamined. n490 Rather, what was striking was that the Court proceeded by first determining that, right or wrong, the Miranda Court had expressed its ruling as one it thought was grounded in the Constitution n491 and then concluding that Congress violated the Constitution simply by flouting the Court's word. n492
The 1992 joint opinion of Justices O'Connor, Kennedy, and Souter in Planned Parenthood of Southeastern Pennsylvania v. Casey, n493 reaffirming the core of Roe v. Wade, n494 offers a final example of the Court's remarkable hubris. Though I admire most of that opinion as one of the Court's most eloquent explications of liberty and equality, its concluding passages reveal the Justices' unfortunate impatience with dissent from their wisdom when they announce that retreating, especially in the face of continued protest, from a decision as basic as Roe would put in doubt the august tribunal's legitimacy as the oracle of last resort on fundamental national matters. n495 Indeed, the Court [*290] insisted, even more is at stake: "If the Court's legitimacy should be undermined, then, so would the country be in its very ability to see itself through its constitutional ideals. The Court's concern with legitimacy is not for the sake of the Court, but for the sake of the Nation to which it is responsible." n496 Evidently, far from being the "least dangerous" branch, the Court turns out to be the one truly indispensable branch; without it, sadly, we would have no one to define our "constitutional ideals" for us. The Court's frame of mind seems aptly captured in Justice Kennedy's observation to a reporter in his chambers just before the Court announced its decision in Casey, when he said it was hard for a Justice in his position to know "if you're Caesar about to cross the Rubicon or Captain Queeg cutting your own tow line." n497
Although Chief Justice Rehnquist and Justices Scalia and Thomas played Brutus regarding the antidemocratic substance of the opinion in Casey, they have had no qualms about assuming the role of Caesar on other occasions. Indeed, they joined two of Casey's three authors (Justices O'Connor and Kennedy) in all but one n498 of the cases invoked above for the proposition that the Justices in the Bush v. Gore majority have little but disdain for Congress as a serious partner in the constitutional enterprise, and not much patience with "We the People" as the ultimate source of sovereignty in this republic. n499
C. Creating Constitutional Dialogue
Given that set of attitudes, the Court's sense that its destiny was to intervene - to take a hit to its short-term reputation, if necessary - need not have been driven by any genuine fear that the nation was about to hurtle over some constitutional precipice. For the Court, the [*291] spectacle of perhaps another week of state and local officials holding ballots up to the light, followed by the likely display of unabashed partisanship on Capitol Hill, provided incentive enough to intervene. Those images, which struck many as a display of democracy in motion, were obviously the very antithesis of the rule of law to the Bush v. Gore majority. To those for whom, in Justice Scalia's phrase, "the rule of law is the law of rules," n500 the fact that the rules laid down in this case - by the Twelfth Amendment, by Congress in 1887, and by the Florida Legislature as its words had been construed by Florida's highest court - would appear to have invited this boisterous political brawl must have seemed an oversight by the Framers, deserving no greater respect than a gigantic ink blot. The institutional incentives were thus aligned powerfully in favor of decisive Supreme Court intervention to bring closure to an election that had already overstayed its welcome as the Christmas season neared - even if that meant distrusting the people and insulting democracy. For a Court that believes its own press, having absorbed the television image of a tribunal above the fray that alone can speak with authority about the Constitution, nothing could have seemed more natural.
But there is another vision of the Court's proper role in a democracy, a vision in which the order the Court sought by avoiding politics is instead the work of politics, in which order is earned rather than decreed. When the Court proclaimed the unconstitutionality of racial segregation in public schools, n501 it entered into a still-ongoing dialogue with the political branches, and particularly with Congress, n502 recognizing that a sweeping declaration of principle was all the nation could handle in a single gulp. And when the Court ordered President Nixon to turn over the tapes that the Watergate special prosecutor sought for use in a pending criminal trial, n503 it was engaged in facilitating, not circumventing, the ability of the political branches - the Executive, in pursuing a criminal prosecution, and Congress, in conducting impeachment hearings - to carry on their constitutionally defined roles. Come to think of it, even Marbury v. Madison, n504 the Mother Ship in the judicial review fleet, did its work not by displacing the actions of the political branches within their own spheres, and not even by ordering anyone to do anything, but by diplomatically declining, as beyond its constitutional authority to receive, the jurisdiction it construed Congress as having attempted to confer upon it. n505
[*292] In this other vision of the Court's proper role vis-a-vis political action, Congress and the Executive are less to be feared than challenged; each may or may not rise to the occasion, with ordinary politicians gaining for themselves in an extraordinary time some of the respect that they have not always earned when their task was less daunting. Who knows but that a kind of bipartisanship appropriate to a virtually tied election and to a challenge of historic proportions might not have developed in the House and Senate? n506 And, even if it did not, extraordinary challenges by definition run the risk of failure. Intervening to short-circuit the political process, the Court stands revealed, wielding naked power clothed in the trappings of judicial authority.
VI. Supreme Distraction: The Court's Alleged Extraconstitutional Motive
As unfortunate and worthy of public criticism as these manifestations of the Court's Herculean pretension are, some accuse the Court of an even baser motive: partisan zeal and crude self-interest. Commentators who attack Bush v. Gore not for what the Court did, but for the shamefully partisan purposes those commentators believe they can prove impelled the majority, n507 confront an initial difficulty: two moderate Justices, whom no one could plausibly suspect of being foolish enough to contribute unwittingly to a partisan conspiracy nor corrupt enough to go along knowingly, accepted the equal protection theory adopted by that majority. n508 None of these commentators so much as [*293] acknowledges this difficulty forthrightly, much less suggests a way to overcome it. n509
Beyond that logical obstacle to accepting the improper motive claim lies an equally fundamental problem of proof. Often, the evidence offered to support the theory either is too offhand and anecdotal to take seriously or is logically incoherent in its inception. In the latter category is "evidence" that takes the form of a supposed inconsistency between how a Justice voted on the stay in Bush v. Gore and how that Justice voted on stay applications in death penalty cases in the past, or a supposed irreconcilability between how a Justice voted on the merits in Bush v. Gore and how that Justice voted on allegedly comparable issues in prior cases.
[*294] But it would be fatuous to deem the December 12 deadline - the only issue on which the conservative majority did not have company among the moderates - comparable to any issue the Court had ever confronted. Moreover, the entire strategy of inferring from such alleged inconsistencies an evil scheme to install George W. Bush in office seems ill conceived, for it rests on a fairly transparent fallacy. If someone is not very smart or thinks he is not being watched very closely, then a departure from his usual modus operandi may be a sign that he is up to no good. But if someone is extremely clever and knows that an army of equally clever observers will analyze and dissect his every step from the moment he acts until historians cease to care, then a departure from his normal modus operandi is much less likely to indicate that he is up to no good. Indeed, what is most interesting about the supposedly partisan motivation underlying Bush v. Gore is that it is simultaneously much too obvious to be plausible as an explanation - only the klutziest of closet partisans would expect to get away with stealing an election based on phony and easily refuted reasoning for stopping a recount - and not nearly obvious enough to render the decision so manifestly illegitimate that some form of defiance would be warranted.
That the Justices in the Bush v. Gore majority acted very much in accord with their previously manifested self-aggrandizing proclivities and with their well-established tendency to define equality in surface-hugging and appearance-embracing terms is of course consistent with the possibility that their action was overdetermined, a product both of those proclivities and of a partisan preference for candidate Bush - a preference the Justices in the majority presumably shared, but one that I believe played an ultimately unknowable role in motivating their actions in this case. Thus, while I criticize the suggestion that Bush v. Gore represented a departure, I also challenge the premise that such a departure, if it occurred, could in any event constitute evidence of partisan motive, n510 arguments to the contrary by otherwise sensible observers notwithstanding. n511
[*295] Setting aside the weakness of the evidence offered to prove it, what if we were simply to ask ourselves: do we believe - based on our informed hunches - that the majority Justices' preference for a Bush presidency played a decisive role in how they perceived the events in Florida, how they read the rulings of the Florida Supreme Court and the provisions of Florida law, and how willing they were to take the case up themselves rather than permit political processes to run their course? How likely do we think it is, in other words, that given the many other factors and forces shaping their perceptions and inclinations, they would have interrupted the political process, rather than let the counting continue if they had believed that their intervention would have put Gore rather than Bush in the White House?
The very symmetry that the question posits is misleading. Perhaps a more illuminating inversion would be one that assumes a Gore victory in Florida's popular vote followed by a Bush attempt to contest that result and to obtain a statewide manual recount, culminating in a Florida Supreme Court rebuff of that attempt that left Gore the winner unless the Court were to intervene on behalf of Bush. How likely do we think it is that the Court would have intervened then if a ruling for Bush in that scenario would have required as much - well, call it "creativity" - as the ruling for Bush required in Bush v. Gore?
Posing the question that way separates the perceived crisis-averting, stability-restoring function of the Court's intervention from its supposed Bush-electing function. Those who think that the Court intervened in order to protect Bush should have second thoughts if they doubt the Court would have intervened in order to elect Bush in this alternate scenario. Reinforcing such second thoughts, it is worth recalling that, as matters stood in late November and early December 2000, the Justices in the majority had good reason to suppose that not intervening would probably have led in the end to the same outcome. n512 But even if we assume that these Justices thought their actions would make the difference between President George W. Bush and President Al Gore, I think the question of causation here is genuinely unanswerable and not merely plagued by problems of proof. n513
[*296] Certainly, the post-Bush v. Gore protestations publicly made by some of the Justices in the majority who asserted that partisan political considerations never enter the Court's internal deliberations and made no difference in this instance provide no reason to believe that partisan preferences truly played no role here. Who would expect any Justice who recognized the incompatibility between the office he or she holds and partisan motives in its discharge, to admit those motives privately, much less confess them publicly? For the same reason, the assurances offered by some of the dissenting Justices, who suggested that their colleagues were guilty at most of a terrible misjudgment and not of partisan abuse of power, seem lacking in probative force. The dissenters too, after all - given the power of cognitive dissonance - need to believe that the people with whom they serve and the institution in which they work have not acted to dishonor themselves. Keep in mind that the accusation that the most strident commentators make against the Justices in the majority involves no small infraction: if the majority Justices deliberately distorted their ruling to see their favorite candidate elected and increase the chances of seeing their legacies perpetuated [*297] by the appointment of like-minded jurists, then they violated their oaths of office and are guilty of high crimes and misdemeanors, meriting impeachment, conviction, and removal from office. Who would want to serve with, much less be, someone who deserves that sort of ignominy?
Given the magnitude of the wrong that the conspiracy theorists attribute to the Bush Five and the implausibility of finding credible evidence to support those theories in the statements or writings of any of the Justices, does it make sense to dig deeper to find the "Truth of the Matter"? I doubt it. It is far from clear that a purpose so deeply embedded in the psyches of the judges who hand down a given decision is even relevant, much less dispositive, to the verdict that legal and political critics and historians, as well as the general public, ought to render upon that judicial performance. n514 Not even certainty that the Justices who decided Brown v. Board of Education n515 had shockingly discreditable motives (something I have no reason to believe) could dethrone that holding from its place of honor, just as certainty that Dred Scott's n516 authors were pure of heart could not lift that decision from its well-deserved obloquy.
Beyond its dubious relevance, I think there simply is no "Truth of the Matter" to be unearthed. The point is epistemological as well as psychological, something akin to the way observations in quantum theory alter the phenomena observed, but applied in the admittedly different realm of discourse about judicial motivation. Imagine any procedure for acquiring "knowledge" about precisely what role the Justices' presidential preferences, their intentions regarding retirement, their hopes and fears regarding their possible successors and their legacies (all discounted by how suspicions of partisan motivation might affect those same legacies) played in their decision. Surely, applying any such procedure would affect those states of mind and how they might be expected to operate. The process of observation would, as Heisenberg taught about the physical world, alter the thing observed. Thus, we would have to confine ourselves to extrinsic, preexisting evidence - evidence that would at best provide secondhand insight into psychological processes far too complex to leave a convincing trail of clues to a decisive, but-for trigger for any one Justice's actions, let alone those of a group of five. It therefore makes little sense to talk as though we could, even in principle, know whether partisan considerations really provided the final motivation for the majority's actions in [*298] Bush v. Gore. In the end, asking whether putting Gore's shoe on Bush's foot would have changed the election n517 can reveal little beyond the fact that neither candidate would then be wearing matching shoes.
In addition to the epistemological problem, it must also be kept in mind that, when we are talking not of something relatively straightforward like bribery, but of something as complex and elusive as partisan bias, the danger is great that the bias will be as much in the eye of the accuser as in the heart of the accused. And the wound inflicted on the credibility and legitimacy of the entire Court on which the accused sits may be wholly disproportionate to the force of the evidence adduced. Particularly when the accuser's proclaimed motive is to restore a judicial institution's tarnished reputation so that it may better serve the nation, n518 that wound is not lightly to be ignored.
This brief meditation on the possible relevance of the motive inquiry may have a valuable spin-off: when a judicial decision spawns a fusillade of attacks on its allegedly partisan motivation, when the assaults show no indication of diminishing over time, and when virtually all the attackers and those who find them persuasive are on the losing side of the decision's political impact, that should serve as a clue that the nature of the question was one in which legal and political considerations were so inextricably intertwined that any attempt to render a legal judgment that was both untainted by partisan politics, and perceived to be untainted, was likely to fail. Given this difficulty, the Court should have trusted itself less and the people more, realizing that, although the eventual winners might believe that it acted according to law rather than partisan desire, the eventual losers would likely harbor an enduring distrust of the tribunal in which they once placed so much hope.
The Court expressed sensitivity to such concerns in the plurality opinion in Casey, n519 but left them behind in Bush v. Gore. Had there been no textual basis for saying that the Constitution committed the decision in question to the political branches, one might have argued that the Court was duty bound to plunge ahead despite the danger signs. Even in such a case, much can be said in support of the "passive virtues," n520 of which no less a precedent than Marbury v. Madison remains a brilliant reminder. n521 But there is, as we have seen, just such a textual basis in the language of the Twelfth Amendment and the process it creates for congressional resolution of disputes among [*299] slates of electors who all claim to represent the same state. n522 In these circumstances, the indeterminacy of the attack on motivation and the predictable way in which it has polarized the polity at once render that kind of attack impotent as a means of preaching to anyone but the converted and reinforce the arguments against the Court's having entered the thicket at all.
In the end, the proponents of "bad motive" theories discredit more substantive criticisms of Bush v. Gore through guilt by association. For whenever an attack on a target misfires, the belief that the target is undeserving of attack is reinforced. And that belief, in this case, would be tragically wrong.
VII. The Exceptionality of Bush v. Gore
It would be easy to confuse a refusal to condemn the Court in an ad hominem and ad feminam way with a willingness to treat Bush v. Gore as just another mistaken decision. But the character of the mistake renders the case unique. The Court was, after all, deciding the outcome of a presidential election in which the Constitution's text pointed to a more overtly political decisionmaker - Congress - and in which the Court's seizure of authority from that decisionmaker would, in all likelihood, be perceived as shaping the future of the Court's own composition, and thus of the law, for decades to come. Given how breathtakingly strange it seemed that the Court would willingly accept that apparent conflict of interest and commandeer a presidential election, not one major constitutional scholar - not even those of us who for several years had been criticizing the Court for its arrogance with respect to matters constitutional - predicted before the Court granted certiorari on November 24 that it would intervene. We were all positive that the Court would sit this one out.
That was not to be. Instead, the Justices took their "unsought responsibility" to a new level and, in so doing, did more than simply defy our predictions. By plunging ahead against all reason, and by rationalizing its action in such pathetically weak terms, the Court dishonored a legal tradition to which most of us had devoted our lives, leading many of us to question, at least momentarily, our long-held "constitutional faith." n523
But the passage of time has assuaged that deep disappointment and has helped us recall that "the Court" is much more than the nine Justices who sit on it at any one time. In fact, from the perspective of a teacher of constitutional law, the ruling serves as a tremendously illuminating lens through which to examine the intersection of numerous [*300] lines of doctrine and to explore the character of the current Court. Its notion that a political crisis was afoot, that the crisis touched our constitutional order, and that, therefore, only the Court could come to the rescue, was of a piece with so much else this Court has done of late that Bush v. Gore, for all its drama and absurdity, has become sadly easy to teach.
This is not to say that I regard the trajectory of judicial attitudes and approaches of which Bush v. Gore is a kind of apotheosis as somehow predestined, with an indefinite series of future decisions in aggressively conservative directions taken as a given. On the contrary, the eventual meaning of Bush v. Gore could depend on how the presidency of George W. Bush - for which the current Court will, rightly or wrongly, be given credit or blame in the public mind - unfolds: on the judicial nominations this Republican President makes and how the currently Democratic Senate responds, on the midterm elections of 2002 and the presidential election of 2004, on the kinds of legislative reforms of the election process that the experience of 2000 triggers. It could also depend on how the Supreme Court, as reconfigured by the next several appointments, treats Bush v. Gore. The Court could treat the decision as a taking-off point for a meaningful extension of equal protection doctrine from the realm of recounts to the realm of how votes are counted in the first place, and thus prevent it from remaining arbitrarily restricted to the context that spawned it. n524 Or the Court could treat Bush v. Gore as an object lesson counseling against judicial "activism" in those cases in which .
My own favorite filler for the blank in the preceding paragraph would use Bush v. Gore to counsel against judicial activism in cases in which such textual and structural guidance as the Constitution provides points toward permitting the political branches to resolve the dispute under procedures at least sketched, if not fully spelled out, by the Constitution itself and by statutes or treaties promulgated pursuant to it; in which there is no clearly threatened violation of any constitutional right that the political branches are structurally incapable of, or indisposed toward, protecting; in which taking jurisdiction away from the political branches in the circumstances presented would yield no coherent remedy for whatever right is said to be threatened; in which the contemplated judicial action would serve to entrench the power of the political party or group seeking such action rather than to protect relatively powerless individuals or groups from the entrenched power of others; or in which the judicial action in question would advance [*301] the interests of the controlling majority on the Court in some extrinsic way, as by assuring the nomination of like-minded replacements. Bush v. Gore fails on all five counts.
Alternative approaches to filling in the blank could range from foolish non sequiturs to tragic extrapolations. The counsel against activism could be universal, suggesting a "go slow" posture on essentially everything. The counsel could be politically polarized, suggesting passivity with respect to either "liberal" or "conservative" causes. Or it could be tied to whole chunks of the constitutional architecture, urging passivity, or even total withdrawal, with respect to structural matters such as those bearing on the separation of powers and/or federalism or with respect to all rights whose derivation entails going beyond the freedoms or privileges expressly enumerated in the Constitution's text and reflected clearly in its history, such as those rights that rely on structural or philosophical inferences, whether called penumbras (usually by liberals) or postulates (usually by conservatives).
The idea of Bush v. Gore as a kind of "negative object lesson," which seems to me more plausible than the idea of Bush v. Gore as the seed for some wonderfully flowering tree of doctrine that reveals the Constitution in a yet brighter light, brings to mind an occasion several years ago when my wife and I were having dinner in a Chinese restaurant near Cambridge's Central Square. After dinner, we were quite elegantly served two fortune cookies, each on its own oval dish. After opening the first, which we had agreed in advance to treat as applicable to us both, we decided not to open the second. We saved the "fortune" that cookie contained: "May your life serve as an example to others how not to lead theirs." Gulp. From time to time, we have been tempted to give the fortune away, but we could never find quite the right recipient. That problem was solved for us by Bush v. Gore.
Although I think I have gained some perspective on the decision in my role as a teacher of constitutional law, appearing before the Court in the wake of Bush v. Gore has been another matter. I did so twice within the several months following the decision. n525 The sense of solidarity, of shared pride, that I used to feel each time I got up to argue in that marble building - now more than thirty times in all - seemed to have slipped away, replaced by a grim realism of a sort to which I have not grown, and hope never to grow, entirely accustomed. In the past, I had disagreed fairly often with decisions the Court had rendered, sometimes in cases I had argued (most notably, Bowers v. [*302] Hardwick n526 and Rust v. Sullivan, n527 as well as Timmons v. Twin Cities Area New Party n528), but those disagreements had not left me with a strong feeling of disconnection, of alienation from a core enterprise that, in significant part, defines the current Court and its sense of its constitutional role.
Understanding even more clearly the nature of the beast, as it were, could comfort me as an academic, but offered cold comfort indeed each time I joined in its enterprise by entering its arena as gladiator rather than as spectator. Thus, to the question whether I see the errors the Court committed in Bush v. Gore as sui generis, I am constrained to reply, no, not sui generis from the perspective of an observer and critic; but yes, perhaps sui generis from the perspective of a participant and virtual co-conspirator.
Cognitive dissonance has no doubt already done its work both on the dissenting Justices and, less importantly, on me. Who, after all, would want to be part of, or to argue in front of, a tribunal whose vision of politics and of the imperative of democratic participation is so stunted? So one gradually accepts excuses, explanations. The tribunal's vision was stunted, yes, but the circumstances were so exigent, so confusing, the advocacy so compromised by the political matrix from which it arose. Maybe. But once all the excuses have been set forth and assimilated, there is likely to remain, for me at least, a sad lump in the throat that will, stubbornly, never go away.
VIII. Lessons for the Future
To draw a reason to doubt the majesty of law and of the Constitution from the Court's contented assurance that it, and it alone, could rise to this constitutional occasion would be tragically misguided. That the Court's distressing decision to rescue democracy from itself was no isolated misstep, but a manifestation of pathologies evidenced in all too many other ways, is hardly a basis to diagnose either judicial review or the rule of law as moribund or even mortally wounded. The Court has had far better days and will have them again. But drawing sweeping institutional lessons in any direction from Bush v. Gore and from the political resolution it denied the people seems dangerous.
The impulse to overreact to legal moments like Bush v. Gore is strong - but the best interests of the people and their constitutional institutions demand a longer perspective. I especially lament the tendency [*303] of some to draw from the Herculean pretension that marked Bush v. Gore the lesson that judicial modesty and incrementalism are always the path of wisdom, that the Court that decides the least decides the best. n529 Other commentators will doubtless use the decision to reach beyond even the misguided call for blanket judicial self-restraint and argue that - now more than ever - the wings of judicial review must be decisively clipped. n530 This response draws precisely the wrong lessons from the case. A healthy constitutional pluralism - relaxing the Court's jealous grip on all matters potentially constitutional - is a far cry from the sort of debilitating constitutional populism that relegates the Court to second-class status. Judicial review must not be rendered toothless as a forceful check on democratic excess. Nor should temporary dissatisfaction with the Court lead to a kind of institutional isolationism, leaving each governmental branch with the constitutional power only to resist encroachments upon its own turf.
Movements like these would strike at the heart of Marbury v. Madison, a decision that displays the Court's capacity to get it right even in the face of political turmoil, whether genuine or media-manufactured. In many ways, Marbury is the mirror image of mirror-laden Bush v. Gore. With exquisitely self-conscious sensitivity, the Marbury Court chose the perfect legal vehicle to establish enduring principles without undermining any aspect of the democratic political process. In the world through the looking glass, the Bush Court felt compelled to intervene in the dispute of the moment, stumbling upon a vehicle that undermined the political process without an enduring principle in sight.
The mirror shows us what Marbury might have looked like with a Rehnquist regime in place: "Section 13 of the Judiciary Act of 1789 confers jurisdiction upon this Court to issue the requested writ. Madison's argument that Section 13 exceeded Congress's powers is interesting but lacks force inasmuch as we do not sit as a Council of Revision to pass on the validity of congressional enactments. No power of judicial review having been conferred in the Constitution's text, we may assert no such power. Hence the writ shall issue" - at which point all hell breaks loose, Jefferson tells Madison to defy the Court, Marshall and several colleagues are impeached and convicted, and the next 200 years look entirely different.
[*304] What the Court needs now is not a curtailment of the power that Marbury established, but a return to the contextual self-awareness that Marbury displayed. How much and when the Court should decide depends on the constitutional principle to be vindicated, the political setting in which a controversy is embedded, and the social, cultural, and historical forces at play. And the arguments over when the Court does well to act more boldly - when the time and circumstances are right for a Brown v. Board of Education, a Gideon v. Wainwright, n531 a New York Times v. Sullivan, n532 a Baker v. Carr, n533 or a Roe v. Wade - are too complex and multifaceted to be squeezed into any simplistic generalization that would treat Bush v. Gore as the right wing's answer to Roe v. Wade. To use Bush v. Gore to eviscerate Marbury and the host of justly celebrated decisions in Marbury's mold seems to me perverse in the extreme. The danger of always fighting the last war is too real for us to succumb to any version of the syndrome that would make Bush v. Gore our judicial Vietnam.
With that in mind, the main lesson I would draw from Bush v. Gore is addressed more to future Courts than to this one: When a generation comes to one of those rare moments in which a confluence of legal and political factors creates a challenge of historic proportion - I say a challenge, not a crisis, for crisis there was none - the Court should take a long, hard look in the mirror of its own past and at the past of the politics we have lost and recognize in that challenge opportunities for both political and intellectual greatness, opportunities in which the other branches, if the matter is one best understood as left to them by the Constitution, may, without rescue by the Court, summon the better angels of our nature to the work of achieving a "more perfect Union."
FOOTNOTES:
n1. See Richard I. Aaron et al., Law Professors Denounce Supreme Court Ruling, at http://www.the-rule-of-law.com (last visited Oct. 7, 2001).
n2. E.g., Vincent Bugliosi, The Betrayal of America: How the Supreme Court Undermined the Constitution and Chose Our President (2001); Alan M. Dershowitz, Supreme Injustice: How the High Court Hijacked Election 2000 (2001).
n3. E.g., The Vote: Bush, Gore, and the Supreme Court (Cass R. Sunstein & Richard A. Epstein eds., 2001) [hereinafter The Vote].
n4. E.g., A Badly Flawed Election: Debating Bush v. Gore, the Supreme Court, and American Democracy (Ronald M. Dworkin ed., forthcoming Aug. 2002) [hereinafter A Badly Flawed Election]; Bush v. Gore: The Question of Legitimacy (Bruce Ackerman ed., forthcoming 2001) [hereinafter The Question of Legitimacy].
n5. Fla. Stat. ch. 102.168(3)(c) (2000).
n6. I served as counsel to Vice President Gore in the lawsuit brought by then-Governor Bush in the federal courts to enjoin the recounts in Florida and as his counsel of record in all of the U.S. Supreme Court proceedings.
n7. Judge Richard Posner has argued that the Court's intervention in Bush v. Gore was necessary in part to ensure that the crisis would be decided before Christmas. Pamela Karlan & Richard Posner, Forum - The Triumph of Expedience: How America Lost the Election to the Courts, Harper's, May 2001, at 31, 32.
n8. N. Sec. Co. v. United States, 193 U.S. 197, 400-01 (1904) (Holmes, J., dissenting).
n9. 121 S. Ct. 525 (2000).
n10. Laurence H. Tribe, Let the Courts Decide, N.Y. Times, Nov. 12, 2000, 4, at 15.
n11. Id.
n12. See Ronald A. Klain & Jeremy B. Bash, The Labor of Sisyphus: The Gore Recount Perspective, in Overtime! The Election 2000 Thriller 157, 159-60 (Larry J. Sabato ed., 2001).
n13. For a refreshing exception, see David A. Kaplan, The Accidental President 87-92 (forthcoming Oct. 2001), which notes that "the Bush campaign cast the first legal stone between the candidates," id. at 90, and that "federal court was going to be the GOP grail to the White House," id. at 87.
n14. U.S. Const. art. II, 1, cl. 2 ("Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors, equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress ... .").
n15. The "safe harbor" provision - whose label has stuck since I first affixed it in November 2000 - promises that Congress will regard as presumptively valid the electoral slate provided by a State that completes its election contest proceedings at least six days before the meeting of the electors, a date that 3 U.S.C. 7 fixes as the "first Monday after the second Wednesday in December." 3 U.S.C. 7 (1994). In this case, the electors were to meet on December 18. Thus, for Florida to obtain the benefit of the safe harbor presumption, all election contests, and any ensuing recounts, would have to be completed by December 12. See id. 5 ("If any State shall have provided, by laws enacted prior to the day fixed for the appointment of the electors, for its final determination of any controversy or contest concerning the appointment of all or any of the electors of such State, by judicial or other methods or procedures, and such determination shall have been made at least six days before the time fixed for the meeting of the electors, such determination ... shall be conclusive, and shall govern in the counting of the electoral votes as provided in the Constitution, ... so far as the ascertainment of the electors appointed by such State is concerned.").
n16. See Brief for Appellant at 12-45, Pennzoil Co. v. Texaco, Inc., 481 U.S. 1 (1987) (No. 85-1798); Reply Brief for Appellant at 3-11, 13-15, Pennzoil, 481 U.S. 1 (1987) (No. 85-1798).
n17. Pennzoil, 481 U.S. at 3.
n18. Touchston v. McDermott, 120 F. Supp. 2d 1055, 1055-56 (M.D. Fla. 2000); Siegel v. LePore, 120 F. Supp. 2d 1041, 1044 (S.D. Fla. 2000).
n19. Touchston v. McDermott, 234 F.3d 1133, 1134 (11th Cir. 2000) (finding no abuse of discretion in the district court's refusal to enjoin the recounts); Siegel v. LePore, 234 F.3d 1163, 1179 (11th Cir. 2000) (same).
n20. Touchston v. McDermott, 121 S. Ct. 749 (2001) (mem.).
n21. Texaco won at the district court level, Texaco, Inc. v. Pennzoil Co., 626 F. Supp. 250, 262 (S.D.N.Y. 1986), and in the Second Circuit Court of Appeals, Texaco, Inc. v. Pennzoil Co., 784 F.2d 1133, 1157 (2d Cir. 1986), even though it lost in the Supreme Court, Pennzoil, 481 U.S. at 3 (1987).
n22. 481 U.S. 1 (1987).
n23. The credit for the strategy and tactics pursued before the county canvassing boards and in the state courts in the Florida vote-counting contest in fact belongs not solely to David Boies, brilliant strategist and tactician though he is, but also, and perhaps principally, to Ronald A. Klain, who had been Vice President Gore's extraordinary Chief of Staff in the White House and was his chief legal counselor during the election and in the recount litigation. For an authoritative and unabashedly partisan yet eminently fair account of the legal and political thinking that drove the post-November 7 Gore strategy, see Klain & Bash, supra note 12, at 160-63.
n24. The common assumption that the Supreme Court's decision at least allowed Florida to fall within the safe harbor is incorrect. See infra pp. 267-68.
n25. NCAA v. Tarkanian, 488 U.S. 179, 193 (1988).
n26. U.S. Const. art. II, 1, cl. 2.
n27. See U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 783-84 (1995) (invalidating a provision of the Arkansas Constitution that prohibited incumbents who had served in a given office more than a certain number of terms from having their names placed on the ballot for that office). Justice Stevens's opinion for the Court, which Justices Kennedy, Souter, Ginsburg and Breyer joined, treated a state's U.S. senators and representatives as fully federal officers and rejected the petitioners' contention that the "original powers of sovereignty that the Tenth Amendment reserved to the States" included the power to fix the terms of their own members of Congress. Id. at 802.
n28. See, e.g., Bouie v. City of Columbia, 378 U.S. 347, 353-54 (1964) (holding that a state supreme court's interpretation of state criminal law that deviates entirely from state precedents violates the ex post facto principles embodied in the Due Process Clause); NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 454-58 (1958) (holding that a state supreme court's finding that individuals were procedurally barred under state law from bringing constitutional claims did not foreclose federal court review of those claims).
n29. See U.S. Term Limits, 514 U.S. at 845 (Thomas, J., dissenting). Chief Justice Rehnquist and Justices Scalia and O'Connor joined Justice Thomas's dissent in U.S. Term Limits, which contended that the states at the time of the Constitutional Convention reserved the power to define the qualifications of their own representatives in Congress. Id. at 865, 886-88. It would seem that the power to allocate decisionmaking authority among the states' own governmental organs, even with respect to such federal officers as presidential electors, would be at least as fundamental an attribute of a state's sovereignty as the power to select the qualifications of its federal representatives.
n30. Even the advocates for Vice President Gore did not endorse such an extreme reading of Article II. Rather than assert that Article II gives the Supreme Court no authority to intervene, our brief argued that the Florida Supreme Court's rulings did not raise Article II issues in this particular case. Brief of Respondent Albert Gore, Jr. at 11-22, Bush, 121 S. Ct. 525 (2000) (No. 00-949), available at 2000 WL 1809151, at 11-22.
n31. Bush, 121 S. Ct. at 529 (deciding the case on equal protection grounds and thus not overturning the Florida Supreme Court's interpretation of Florida law); Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471, 475 (2000) (remanding the case to the Florida Supreme Court on the grounds that the Court was "unclear as to the extent to which the Florida Supreme Court saw the Florida Constitution as circumscribing the legislature's authority under Art. II, 1, cl. 2 ... [and] as to the consideration the [court] accorded to 3 U.S.C. 5").
n32. See Bush, 121 S. Ct. at 534-37 (Rehnquist, C.J., concurring).
n33. In hindsight there is strong reason to doubt that, if asked, the Florida Supreme Court would have read the Florida statutes to require the recount to end by December 12. See Gore v. Harris, 773 So. 2d 524, 528-29 (2000) (Shaw, J., concurring). The part of the state court decisions of November 21 and December 11 to which the Bush v. Gore per curiam referred in declaring that all recounts had to be completed by December 12, see Bush, 121 S. Ct. at 533 ("The Supreme Court of Florida has said that the legislature intended the State's electors to "participate fully in the federal electoral process,' as provided in 3 U.S.C. 5." (quoting Gore v. Harris, No. SC00-2431, slip op. at 27 (Fla. Dec. 8, 2000))), dealt not with deadlines for the recounts, but instead with the limitations on the power of the Secretary of State to reject late returns from county canvassing boards. See Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1273, 1289 (Fla. 2000) ("We conclude that the reasoned basis for the exercise of the Department's discretion to ignore amended returns is limited to those instances where failure to ignore the amended returns will ... preclude a candidate, elector, or taxpayer from contesting the certification of an election ... or result in Florida voters not participating fully in the federal electoral process, as provided in 3 U.S.C. 5." (emphasis added)); Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1220, 1237 (Fla. 2000) ("Ignoring the county's returns is a drastic measure and is appropriate only if the returns are submitted to the Department so late that their inclusion will [either] ... preclude a candidate, elector, or taxpayer from contesting the certification of an election ... [or] preclude Florida voters from participating fully in the federal electoral process." (emphasis added) (citing 3 U.S.C. 1-10 (1994))). Neither did the state court's opinion of December 8, which ordered the statewide recount, establish December 12 as a deadline mandated by state law. See Gore v. Harris, 772 So. 2d 1243, 1261 (Fla. 2000) ("The need for prompt resolution and finality is especially critical in presidential elections where there is an outside deadline established by federal law."). Even the Florida court dissenters did not contend that the state election code mandated a December 12 deadline. See id. at 1268 n.30 (Wells, C.J., dissenting) ("As the Supreme Court recently noted, 3 U.S.C. 5 creates a safe harbor provision regarding congressional consideration of a state's electoral votes should all contests and controversies be resolved at least six days prior to December 18, 2000, if made pursuant to the state of the law as it existed on election day. There is no legislative suggestion that the Florida Legislature did not want to take advantage of this safe harbor provision." (emphasis added)); id. at 1272 (Harding, J., dissenting) ("Even if I were to conclude that the appellants' allegations and evidence were sufficient to warrant relief, I do not believe that the rules permit an adequate remedy under the circumstances of this case. This Court, in its prior opinion, and all of the parties agree that election controversies and contests must be finally and conclusively determined by December 12, 2000." (emphasis added)).
n34. See infra section IV.B.4, pp. 263-68.
n35. See U.S. Const. art. II, 1, cl. 4 ("The Congress may determine the Time of chusing the Electors, and the Day on which they shall give their Votes; which Day shall be the same throughout the United States.").
n36. 121 S. Ct. 1693 (2001) (holding that the retroactive application of a state supreme court's abolition of its common law "year-and-a-day" rule, which barred prosecution for murder unless the victim died within that period, did not violate the Due Process Clause because the petitioner could have anticipated the change in the law).
n37. U.S. Const. art. I, 10, cl. 1.
n38. Rogers, 121 S. Ct. at 1697 (quoting Marks v. United States, 430 U.S. 188, 191 (1977)) (internal quotation marks omitted).
n39. See Bush, 121 S. Ct. at 543-45 (Souter, J., dissenting); id. at 549-50 (Ginsburg, J., dissenting).
n40. Nor did any member of the Court in Rogers even so much as cite to Bush v. Gore, despite the obvious relevance of the Article II issue. It is as though Bush v. Gore lies in the background like a 10,000-pound gorilla that the Justices see in their peripheral vision but are hesitant to look straight in the eye.
n41. Although Justice Breyer dissented in Rogers, he did not disagree with the majority's assertion that ex post facto principles apply with less force to courts than to legislatures. See Rogers, 121 S. Ct. at 1710 (Breyer, J., dissenting) ("I agree with the Court's basic approach."). Justice Breyer dissented only from the Court's application of that analysis to the facts in Rogers. Id. at 1711. By contrast, Justice Stevens's decision to join Justice Scalia's dissent makes his agreement with Justice Ginsburg in Bush v. Gore less noteworthy.
n42. Bush, 121 S. Ct. at 549 (Ginsburg, J., dissenting) ("By holding that Article II requires our revision of a state court's construction of state laws in order to protect one organ of the State from another, the Chief Justice contradicts the basic principle that a State may organize itself as it sees fit.").
n43. See Smiley v. Holm, 285 U.S. 355, 368 (1932) (holding that "whether the Governor of the State, through the veto power, shall have a part in the making of state laws is a matter of state polity," not federal constitutional interpretation, even when the state legislature is acting under responsibilities prescribed in Article I, Section 4).
n44. See Bush, 121 S. Ct. at 539 (Stevens, J., dissenting) (observing that Article II, in conferring on state legislatures the duty to choose the manner of selecting presidential electors, "does not create state legislatures out of whole cloth, but rather takes them as they come - as creatures born of, and constrained by, their state constitutions"); id. ("The [State's] legislative power is the supreme authority except as limited by the constitution of the State." (alteration in original) (quoting McPherson v. Blacker, 146 U.S. 1, 25 (1892)) (internal quotation marks omitted)).
n45. Cf. Hawke v. Smith, 253 U.S. 221, 230-31 (1920) (finding that because Article V specifically confers upon state "Legislatures" the power to ratify amendments when Congress proposes that mode of ratification - rather than ratification by state conventions - a state may not interfere with that allocation of power by permitting its citizens to override the state legislature's ratification decision by popular referendum). It would, of course, be a different matter if the state legislature itself were to delegate the choice of electors to the state's citizenry or to the state judiciary. See McPherson, 146 U.S. at 34-35 ("It is, no doubt, competent for the legislature to authorize the governor, or the Supreme Court of the State, or any other agent of its will, to appoint these electors. This power is conferred upon the legislatures of the States by the Constitution of the United States ... ." (quoting S. Rep. No. 43-395 (1874)) (internal quotation marks omitted)).
n46. Even some generic assignments - say, a state constitutional provision that entrusted all election regulations to the state's highest court, sitting as a lawmaking body - would be inconsistent with Article II in the context of choosing the state's presidential electors. See McPherson, 146 U.S. at 39-40 (upholding, against constitutional challenges brought under Article II, Section 1, Clause 2 and the Fourteenth Amendment's Equal Protection Clause, the power of the state legislature to choose a district-based electoral voting scheme). In a passage that was pure dictum in McPherson, the Court quoted with seeming approval a Senate report asserting exaggeratedly that:
The appointment of [presidential] electors is thus placed absolutely and wholly with the legislatures of the several states... . The legislature may provide that they shall be elected by the people of the State at large ... or in districts ... . It is, no doubt, competent for the legislature to authorize the governor, or the Supreme Court of the State, or any other agent of its will, to appoint these electors. This power is conferred upon the legislatures of the States by the Constitution of the United States ... . Whatever provisions may be made by statute, or by the state constitution, to choose electors by the people, there is no doubt of the right of the legislature to resume the power at any time, for it can neither be taken away nor abdicated.
Id. at 34-35 (quoting S. Rep. No. 43-395 (1874)) (internal quotation marks omitted).
n47. Bush, 121 S. Ct. at 537 (Rehnquist, C.J., concurring).
n48. See id. at 539-40 (Stevens, J., dissenting); id. at 545 (Souter, J., dissenting); id. at 549-50 (Ginsburg, J., dissenting); id. at 552-55 (Breyer, J., dissenting).
n49. But this suggestion is seemingly implicit in the questions posed by Kathleen M. Sullivan & Gerald Gunther, Constitutional Law 116-18 (14th ed. 2001). See also id. at 118 ("Was the concurring justices' approach consistent with the position they took in dissent in [U.S.] Term Limits? ... Why wasn't the state free to set forth its own procedures for settling its people's vote - including judicial review of voting disputes - subject only to the limits of its own constitution?"); id. ("Is [Justice Ginsburg's position in Bush v. Gore] consistent with the majority position in Term Limits? If states have sovereignty to control the selection of their presidential electors without federal judicial interference, why not the sovereignty to control the length of their congressional representatives' terms?").
n50. See supra pp. 188-89.
n51. Rogers v. Tennessee, 121 S. Ct. 1693, 1699, 1702 (2001).
n52. Id. at 1702.
n53. The Court in Rogers independently analyzed application of the year-and-a-day rule in prior state court rulings such as Percer v. State, 103 S.W. 780 (Tenn. 1907), and agreeing that the rule had "never once served as a ground of decision in any prosecution for murder," found it not fundamentally unfair to apply the change in the rule retroactively. Rogers, 121 S. Ct. at 1701-03.
n54. Rogers, 121 S. Ct. at 1710 (Breyer, J., dissenting).
n55. If a federal court may supervise a state high court's characterization of state common law for ex post facto purposes, see Rogers, 121 S. Ct. at 1701-03, it may certainly oversee that court's interpretation of state legislation for Article II purposes.
n56. Many of the most sophisticated champions of what the Court did, although not necessarily of what it said, peg the ultimate ruling on Article II. Judge Richard Posner charges that "what the Florida supreme court did with the statute was so freewheeling as to raise a serious question of conformity with Article II of the U.S. Constitution, which places the authority to determine the manner of appointment of a state's Presidential electors in the state's legislature." See Richard A. Posner, Breaking the Deadlock: The 2000 Election, the Constitution, and the Courts 127 (2001); see also Einer Elhauge, Florida's Vote Wasn't "Irregular,' Wall St. J., Nov. 13, 2000, at A36; Charles Fried & Ronald Dworkin, "A Badly Flawed Election': An Exchange, N.Y. Rev. Books, Feb. 22, 2001, at 8.
n57. 501 U.S. 452 (1991).
n58. Id. at 460 ("Through the structure of its government, and the character of those who exercise government authority, a State defines itself as a sovereign.").
n59. Bush, 121 S. Ct. at 537 (Rehnquist, C.J., concurring).
n60. This dissent also quotes Bouie v. City of Columbia, 378 U.S. 347 (1964), which insists that ex post facto principles limit equally the retroactive application of judicial and legislative changes in the law for "if a state legislature is barred by the Ex Post Facto Clause from passing such a law, it must follow that a State Supreme Court is barred by the Due Process Clause from achieving precisely the same result by judicial construction." Id. at 353-54.
n61. Rogers v. Tennessee, 121 S. Ct. 1693, 1704 (2001).
n62. Id.
n63. See Elhauge, supra note 56; Fried & Dworkin, supra note 56.
n64. See, e.g., Robert H. Bork, Sanctimony Serving Politics: The Florida Fiasco, New Criterion, Mar. 2001, at 4, 11 ("Bush v. Gore was a valiant effort, legitimate in law, to rein in runaway political passions and a lawless state court those passions had captured.").
n65. "Undervotes" are ballots on which the machine failed to detect a vote for President. Bush, 121 S. Ct. at 528.
n66. See Posner, supra note 56; Richard A. Posner, Bush v. Gore as Pragmatic Adjudication (manuscript at 18, on file with the Harvard Law School Library), in A Badly Flawed Election, supra note 4 ("The single most arresting feature of the [Election 2000] litigation to a pragmatist is the worst-case scenario that it avoided."). Judge Posner and Robert Bork are certainly not alone among legal commentators in evaluating the Florida Supreme Court's legal reasoning so harshly. See, e.g., Richard Epstein, "In Such Manner as the Legislature Thereof May Direct": The Outcome in Bush v. Gore Defended, 68 U. Chi. L. Rev. 613 (2001); Fried & Dworkin, supra note 56, at 8 (noting that the Florida Supreme Court "refused to take the hint" after the first U.S. Supreme Court decision and that its reasoning "merits at least as much criticism as ... the Supreme Court of the United States").
n67. Thus it is not the case, as some have argued, that the Florida Supreme Court's approach led to a conclusion different from that reached by its conservative critics because the Florida court read the statutes purposively whereas its critics read them textually. See, e.g., Richard H. Pildes, Democracy and Disorder, 68 U. Chi. L. Rev. 695, 710 (2001) (suggesting that the "intent of the voter" standard employed by the Florida Supreme Court was derived from a statutory norm, rather than being derived directly from the text itself). The approaches of the Florida court and its critics, truth be told, would be tough to tell apart. The difference, in all candor, is that the critics seem never to have read the statutory language with sufficient care.
n68. The Party requested recounts in Broward, Miami-Dade, Palm Beach, and Volusia counties. See Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1220, 1225 & n.3 (Fla. 2000). After initial sample recounts indicated that there might be "an error in the vote tabulation which could affect the outcome of the election," Fla. Stat. ch. 102.166(5) (2000), the four counties began, pursuant to the statutory prescription, see id. ch. 102.166(4)(c), to recount all ballots manually. See Palm Beach, 772 So. 2d at 1225.
n69. See Palm Beach, 772 So. 2d at 1226 (noting the trial court's determination that the deadline was "mandatory"). Miami-Dade took no part in the suit. See id. at 1225 n.3.
n70. Deadline for Certification on County Results, Div. of Elections, Fla. Dep't of State, DE 00-10 (Nov. 13, 2000).
n71. Chapter 102.111 of the Florida election code states that returns filed after the seven-day deadline - in this case, after November 14 - "shall be ignored" by the Secretary, but chapter 102.112 states that the Secretary "may ... ignore[]" returns filed after that deadline. Compare Fla. Stat. ch. 102.111 (2000), with id. ch. 102.112. The Florida Legislature has amended chapters 102.111 and 102.112 since the election. 2001 Fla. Sess. Law Serv. 2001-40 (West).
n72. Palm Beach, 772 So. 2d. at 1226 n.5. Examples of "acts of God" include a power outage causing a mechanical failure of vote tabulation equipment. The parallel between language in the Florida Supreme Court's decision in Boardman v. Esteva, 323 So. 2d 259 (Fla. 1975), and that used by the Secretary might make one suspect that the 1975 case was her source. See Boardman, 323 So. 2d at 269 (holding that, in determining whether to invalidate otherwise legal votes, courts should consider "(a) the presence or absence of fraud, gross negligence, or intentional wrongdoing; (b) whether there has been substantial compliance with the essential requirements of the absentee voting law; and (c) whether the irregularities complained of adversely affect the sanctity of the ballot and the integrity of the election"). But any such use of Boardman would turn that decision on its head. Boardman set rigid constraints on the state's ability to invalidate otherwise legal votes. See infra pp. 206-07 and notes 128-132. To turn the shield Boardman provided for the right to vote into a sword to cast aside tens of thousands of ballots cannot be deemed a reasonable - much less legitimate - interpretation of state law.
n73. Palm Beach, 772 So. 2d at 1239. The court first applied traditional canons of statutory construction and determined that "the mandatory language in chapter 102.111 has been supplanted by the permissive language of chapter 102.112." Id. at 1235-36. Relying on the Florida Constitution and previous case law to determine the bounds of the statutorily granted discretion to ignore otherwise valid votes under the "may ignore" provision, the court then concluded that "the authority of the Florida Secretary of State to ignore amended returns submitted by a County Canvassing Board may be lawfully exercised only under limited circumstances." Id. at 1237.
n74. See id. at 1239 (noting that Florida law mandates that "the will of the electors supersedes any technical statutory requirements" (citing State ex. rel. Chappell v. Martinez, 536 So. 2d 1007, 1008-09 (Fla. 1988))).
n75. Id.
n76. See Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471, 475 (2000). The Florida court's November 21 opinion did not explain the choice of November 26 beyond stating a purpose "to allow maximum time for contests pursuant to chapter 102.168," while providing sufficient time to finish county-based manual recounts from the protest phase. Palm Beach, 772 So. 2d at 1240.
n77. A standing federal consent decree reflecting the federal interest in counting overseas ballots (many of which come from members of the armed forces) had extended the Florida statutory deadline for certification of results within seven days of an election by providing that "overseas ballots must be received by counties up until midnight on the tenth day following the election." Palm Beach, 772 So. 2d at 1288.
n78. Judge Posner and others identify the state court's November 26 deadline extension as the worst consequence of that court's November 21 decision. Posner argues that "by moving the boundary between the protest and contest phases, the court squeezed the contest phase virtually to death." Posner, supra note 56, at 96. Professor Michael McConnell argues that "the court added twelve days to a phase that had no real legal significance, while shortening the time for obtaining genuine legal relief." Michael W. McConnell, Two-and-a-Half Cheers for Bush v. Gore, 68 U. Chi. L. Rev. 657, 668 (2001). These charges mistakenly assume that the November 26 deadline would have resulted in the filing of a contest outside the time allowed by statute. However, chapter 102.168 allows a contestant to file a complaint initiating a contest up to ten days after the last county canvassing board certifies results. Fla. Stat. ch. 102.168 (2000). Even absent any action by the Florida Supreme Court, the last county canvassing board would have certified official returns on November 18. Chapters 101.5614(8) and 102.168 of the Florida election code would therefore have permitted the initiation of a contest any time before November 28. The practical effect of the Florida Supreme Court's "extension" was only to delay the initiation of the contest until November 27 - a date within the time already permitted by statute for the initiation of a contest.
n79. Cf. Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175 (1989).
n80. Fla. Stat. ch. 102.168(3)(c) (2000). Gore identified a net gain of 168 votes from a partial manual recount in Miami-Dade and a net gain of 215 votes from a full manual recount in Palm Beach County that were not included in the certified total. Gore v. Harris, 772 So. 2d 1243, 1260 (Fla. 2000).
n81. The election results certified by Secretary of State Harris indicated a 537-vote margin in favor of Bush. See Gore, 772 So. 2d at 1247. Vice President Gore pointed to various problems in the vote totals: 215 Gore votes rejected in Palm Beach County; 168 Gore votes rejected in Miami-Dade; 15 votes netted for Bush from ballots counted after Thanksgiving in Nassau County; 3300 votes rejected in Palm Beach, many of which would have gone to Vice President Gore; and 9000 votes tabulated in Miami-Dade County as non-votes by the machine and never manually tabulated. Id. at 1268. The Florida Supreme Court found that the approximately 9000 contested ballots were enough to "place in doubt" the results of the presidential election. Id. at 1260-61.
n82. Chapter 102.168(8) vests the circuit judge hearing the contest with authority to "fashion such orders as he or she deems necessary to ensure that each allegation in the complaint is investigated, examined, or checked, to prevent or correct any alleged wrong, and to provide any relief appropriate under such circumstances." Fla. Stat. ch. 102.168(8) (2000) (emphasis added).
n83. Gore, 772 So. 2d at 1261.
n84. Id. at 1262.
n85. Id. at 1257.
n86. See id. at 1256. Chapter 101.5614 of the Florida election code, titled "Canvass of returns," explains the procedures for the tabulation of votes. The statute provides that, as soon as the polls are closed, election officials are to "count the number of voted ballots, unused ballots, and spoiled ballots." Fla. Stat. ch. 101.5614(1)(a) (2000). The relevant portion of chapter 101.5614(5) states:
If any ballot card of the type for which the offices and measures are printed directly on the card is damaged or defective so that it cannot properly be counted by the automatic tabulating equipment, a true duplicate copy shall be made of the damaged ballot card ... . or the valid votes on the damaged ballot card may be manually counted at the counting center by the canvassing board, whichever procedure is best suited to the system used. If any paper ballot is damaged or defective so that it cannot be counted properly by the automatic tabulating equipment, the ballot shall be counted manually at the counting center by the canvassing board. The totals for all such ballots or ballot cards counted manually shall be added to the totals for the several precincts or election districts. No vote shall be declared invalid or void if there is a clear indication of the intent of the voter as determined by the canvassing board.
Id. ch. 101.5614(5). The very subject of that provision - instructions for county canvassing boards on how to handle ballots - would seem to make it a logical provision from which to seek legal guidance when doing a manual recount.
n87. Fla. Stat. ch. 101.5614(4) (2000) (emphasis added).
n88. Id. ch. 101.5614(5) (emphasis added).
n89. Id. ch. 101.5614(8) (emphasis added).
n90. Chief Justice Wells actually referred only to the 9000 uncounted ballots in Miami-Dade County, arguing that there was "no basis in this record" to suggest that those votes were damaged or defective. Gore, 772 So. 2d at 1267 (Wells, C.J., dissenting). Chief Justice Wells, however, spent very little of his dissent discussing chapter 101.5614. His only other comment regarding that chapter related to the equal protection argument that would drive the Court's opinion in Bush v. Gore. He stated that, even if the state court majority was correct in contending that chapter 101.5614 "correctly annunciates the standard by which a county canvassing board should judge a questionable ballot," that standard was "fraught with equal protection concerns" because it "utterly failed to provide any meaningful standard." Id.
n91. Id. (quoting Fla. Stat. ch. 101.5614(5) (2000)) (emphasis added).
n92. Chief Justice Rehnquist stated:
There is no basis for reading the Florida statutes as requiring the counting of improperly marked ballots, as an examination of the Florida Supreme Court's textual analysis shows. We will not parse that analysis here, except to note that the principal provision of the Election Code on which it relied, 101.5614(5), was, as the Chief Justice pointed out in his dissent in Gore v. Harris, entirely irrelevant.
Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring) (citation omitted).
n93. Posner, supra note 56, at 97.
n94. 707 So. 2d 720 (Fla. 1998).
n95. Id. at 722 n.4.
n96. Such a definition would also seem to include all overvotes on which the voter's intent was clear. See infra note 260 (defining "overvotes"). Judge Posner argues that the failure of the court "to order the overvotes recounted could not be justified, except possibly by the shortness of time." Posner, supra note 56, at 125. But, under the Florida election code, overvotes are accorded special treatment. Fla. Stat. ch. 101.5614(6) (2000) ("If an elector marks more names than there are persons to be elected to an office or if it is impossible to determine the elector's choice, the elector's ballot shall not be counted for that office, but the ballot shall not be invalidated as to those names which are properly marked."). In fact, Florida Supreme Court Chief Justice Wells argued that, if anything, the recount was flawed because the court failed to order county canvassing officials to check the machine-counted ballots for "legal votes" among those ballots that appeared to indicate the selection of more than one candidate for a single office. See Gore, 772 So. 2d at 1264-65 n.26 (Wells, C.J., dissenting). Presumably, the Florida Supreme Court chose not to order the invalidation of any overvotes sua sponte, neither party having requested a reexamination of any ballot that had already been included in the total. And, as Judge Posner acknowledges, the court may also have been legitimately concerned about the time it would take to review all the machine-counted ballots. Posner, supra note 56, at 106; see also infra p. 267 (discussing interest in meeting the deadline set by the safe harbor provision).
n97. The concurrence refers to Beckstrom in its discussion of the duration of appeals in election contests. Bush, 121 S. Ct. at 538-39 (Rehnquist, C.J., concurring). The parties' briefs and Secretary Harris's letter to the Palm Beach County Canvassing Board also prominently featured the case. See Brief for Petitioners at 2, Bush, 121 S. Ct. 525 (2000) (No. 00-949), available at 2000 WL 1810102, at 2; Brief of Respondent Albert Gore, Jr., supra note 30, at 13, 31-32; Letter from Katherine Harris, Secretary of State, to Palm Beach County Canvassing Board (Nov. 15, 2000), quoted in Palm Beach County Canvassing Board v. Harris, 772 So. 2d 1220, 1227 n.5 (Fla. 2000).
n98. I treat as no support at all in this context Judge Posner's casual reference to a wholly anecdotal newspaper story stating that lower state courts had not been following Beckstrom of late, see Posner, supra note 56, at 107 n.29 (citing Jay Weaver, Law: Check "Defective' Ballots, Miami Herald, Apr. 4, 2001, at 18A) - a claim that, even if true, would hardly mean that the precedent was moribund in the Florida Supreme Court.
n99. See Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring); Posner, supra note 56, at 99.
n100. Transcript of Oral Argument, Bush, 121 S. Ct. 525 (No. 00-949), 2000 WL 1804429, at 57 (Dec. 11, 2000) ("Well, why isn't the standard the one that voters are instructed to follow, for goodness sakes? I mean, it couldn't be clearer. I mean, why don't we go to that standard?").
n101. Bush, 121 S. Ct. at 537 (Rehnquist, C.J., concurring) (citations omitted). To stress his point, the Chief Justice quoted in all capitals the instructions allegedly supplied by counties using punchcard ballots: "AFTER VOTING, CHECK YOUR BALLOT CARD TO BE SURE YOUR VOTING SELECTIONS ARE CLEARLY AND CLEANLY PUNCHED AND THERE ARE NO CHIPS LEFT HANGING ON THE BACK OF THE CARD." Id. (quoting Brief for Respondent at 13 n.5, Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1273 (Fla. 2000) (No. SC00-2346)). But not all punchcard counties provided such precise instructions. In Touchston v. McDermott, 234 F.3d 1133 (11th Cir. 2000), a case brought by voters of Broward County following the election, Judge Tjoflat, dissenting, quoted the instructions found at voting booths in Broward County: "To vote, hold the stylus vertically. Punch the stylus straight down through the ballot card for the candidates or issues of your choice." Id. at 1141 n.19 (Tjoflat, J., dissenting). Even more telling, the provision of the Florida election code to which Chief Justice Rehnquist referred did not require county canvassing boards to provide voters with any instructions on how to cast a ballot, unless the voters specifically requested such assistance. The Chief Justice cited chapter 101.46, Bush, 121 S. Ct. at 537 (Rehnquist, C.J., concurring), but that provision does not require posting instructions at the polling booth. It merely states that "the authorities in charge of elections, where voting machines are used, shall designate suitable and adequate times and places for giving instructions to electors who apply." Fla. Stat. ch. 101.46 (2000). If the Florida Legislature had really intended to make the right to vote dependent on one's ability to read a sign correctly and to accomplish exactly what it requires, regardless of possible obstacles like filled-up receptacles for chads that make it overly difficult to do so, it would presumably have required such instructions at all voting booths.
n102. See John Mintz & Peter Slevin, Human Factor Was at Core of Fiasco, Wash. Post, June 1, 2001, at A1 (noting the lack of Spanish-language ballots in Osceola County).
n103. Posner, supra note 56, at 99 (emphasis added).
n104. Id. at 100 (emphasis added). He also asks how the failure of machines deliberately programmed not to count votes from incompletely punched ballots could possibly be thought an error in tabulation: "If you put a steel bar into a meat grinder and hamburger meat doesn't come out, do you call this an error by the meat grinder?" Id. at 96. Well, yes - if by the "meat grinder" you mean the poor fellow who was busily grinding meat and innocently reached for a steel bar instead. Perhaps more to the point, I would ask: "If you put a circular definition of "legal vote' into a word processor and an argument that isn't question-begging doesn't come out, do you call this an error by the word processor?"
n105. Fla. Stat. ch. 102.141(4) (2000). Florida law mandates an automatic recount whenever the difference in the vote totals between any two candidates for a particular office was less than one half of one percent. Id.
n106. Id. ch. 102.166(4), (5).
n107. Id. ch. 101.5614(5).
n108. See, e.g., Posner, supra note 56, at 99 ("Some of the ballots, however, having been spoiled by the voter, had not been counted (recorded) as votes.").
n109. Id.
n110. Id.
n111. Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring).
N112. Fla. Stat. ch. 102.166(5) (2000).
n113. See Jacobs v. Seminole County Canvassing Bd., 773 So. 2d 519, 522 (Fla. 2000) (per curiam); Taylor v. Martin County Canvassing Bd., 773 So. 2d 517, 519 (Fla. 2000) (per curiam).
n114. This appears to be true, even ignoring the additional irregularities relating to absentee ballots that The New York Times recently documented. See David Barstow & Don Van Natta, Jr., How Bush Took Florida: Mining the Overseas Absentee Vote, N.Y. Times, July 15, 2001, at A1.
n115. See, e.g., Bush, 121 S. Ct. at 537 (Rehnquist, C.J., concurring).
n116. Definition of Error in Vote Tabulation, Div. of Elections, Fla. Dep't of State, DE-00-11 (Nov. 13, 2000). This rigid interpretation is reminiscent of the uncompromising manner in which Harris had construed the oft-mentioned "shall" and "may" portions of the code. See supra note 71.
n117. See Fla. Stat. ch. 97.012(1) (2000) (listing, as one of the responsibilities of the Secretary of State, the duty to "obtain and maintain uniformity in the application, operation, and interpretation of the election laws"); see also Krivanek v. Take Back Tampa Political Comm., 625 So. 2d 840, 842, 845 (Fla. 1993) (refusing to reverse the decision of the county Supervisor of Elections that electors whose names had been temporarily removed from the registration books because they had not voted in several years were not eligible to sign a petition advocating the repeal of a city ordinance); State ex rel. Chappell v. Martinez, 536 So. 2d 1007, 1008-09 (Fla. 1988) (refusing a putative losing congressional candidate's request to order the invalidation of 11,000 votes that the Florida Elections Canvassing Commission had certified).
n118. Fla. Stat. ch. 102.168(8) (2000) (emphasis added).
n119. See generally Johnny C. Burris, The 1989 Survey of Florida Law: Administrative Law, 14 Nova L. Rev. 583 (1990) (surveying Florida appellate court administrative law cases).
n120. Fla. Stat. ch. 102.168(4) (2000) (emphasis added).
n121. Posner, supra note 56, at 120.
n122. Nor does Posner address the court's determination that "the "protest' and "contest' proceedings are distinct proceedings." Gore v. Harris, 772 So. 2d 1243, 1252 (Fla. 2000). A two-decade-old case held that a protest was not even a condition precedent to a contest challenging such a certification after the fact. Flack v. Carter, 392 So. 2d 37, 39 (Fla. Dist. Ct. App. 1980). Flack is flatly inconsistent with Judge Posner's premise that the function of the contest phase is to review rulings by county canvassing boards that occurred in the protest phase, and the Florida Legislature could have addressed this rule had it wished to do so in its intervening amendments to the election code.
n123. See Chevron U.S.A. Inc. v. Natural Res. Def. Council, 467 U.S. 837, 843 (1984); see also 1 Laurence H. Tribe, American Constitutional Law 5-19, at 993-1011 (3d ed. 2000).
n124. Gore, 772 So. 2d at 1260.
n125. In determining whether an agency interpretation of an ambiguous statute is reasonable, courts, of course, examine whether the interpretation is consistent with the legislative purpose and with prior judicial interpretations of the same statute. Burris, supra note 119, at 640 & n.358.
n126. As the Florida Supreme Court stressed in the first phase of the Bush v. Gore litigation, Florida courts should not defer to executive interpretations that are "contrary to law." Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1220, 1228 (Fla. 2000); see also, e.g., Legal Envtl. Assistance Found. v. Bd. of County Comm'rs, 642 So. 2d 1081, 1084 (Fla. 1994) ("When an agency's construction amounts to an unreasonable interpretation, or is clearly erroneous, it cannot stand."); Krivanek v. Take Back Tampa Political Comm., 625 So. 2d 840, 844 (Fla. 1993).
n127. Compare McAlpin v. State ex rel. Avriett, 19 So. 2d 420, 421 (Fla. 1944) (per curiam) (invalidating, in an election for state legislature, ballots that were marked with a "V" instead of an "X" as mandated by statute), and Frink v. State ex rel. Turk, 35 So. 2d 10, 12 (Fla. 1948) (invalidating absentee ballots, cast in a municipal election, for which the electors had incorrectly completed the required affidavits), with Jolley v. Whatley, 60 So. 2d 762, 767 (Fla. 1952) (construing a statute not to require absentee voters to fill out a formal written application and thus refusing to invalidate absentee votes in an election for sheriff).
n128. 323 So. 2d 259 (Fla. 1975).
n129. Id. at 264. Boardman involved absentee ballots, but its "substantial compliance" rule has been extended to other Florida election cases. See State ex rel. Chappell v. Martinez, 536 So. 2d 1007, 1008-09 (Fla. 1988) (relying on Boardman's "substantial compliance" principle in finding that 11,000 votes should not be invalidated although the canvassing board certified them telephonically, rather than in writing as the statute seemed to require).
n130. Boardman, 323 So. 2d at 262.
n131. Id. at 262, 267-68.
n132. See, e.g., Chappell, 536 So. 2d at 1008-09; McLean v. Bellamy, 437 So. 2d 737, 742-46 (Fla. Dist. Ct. App. 1983) (finding no substantial departure from the election code and thus refusing to invalidate absentee ballots cast in an election for the city commissioner that were allegedly defective because, inter alia, the ballots had been mailed out, contrary to statutory requirements, to voters who had not requested them, or only one of the two witnesses that signed the absentee ballot form had physically witnessed the filling out of the ballot); see also Anderson v. Canvassing & Election Bd., 399 So. 2d 1021, 1022-23 (Fla. Dist. Ct. App. 1981) (refusing a candidate's demand that it require the trial court to hold an evidentiary hearing on the validity of ballots that the canvassing board had rejected, and finding that the trial court had enough evidence to apply the Boardman "substantial compliance" standard); Flack v. Carter, 392 So. 2d 37, 40 (Fla. Dist. Ct. App. 1980) (holding that failure to file a protest does not preclude a subsequent contest and ordering the trial court, when it examined ballots in the contest, to consider whether the ballots were in "substantial compliance" with statutory requirements).
n133. See, e.g., Bolden v. Potter, 452 So. 2d 564, 567 (Fla. 1984) (invalidating all absentee ballots because of evidence of vote buying); In re Protest of Election Returns and Absentee Ballots in the November 4, 1997 Election for City of Miami, 707 So. 2d 1170, 1174-75 (Fla. Dist. Ct. App. 1998) (refusing the trial court's remedy of calling for a new election because of uncontradicted statistical evidence of massive fraud in absentee voting, and instead using the traditional common law remedy of invalidating the entire set of absentee ballots). Regarding the butterfly ballot that had misled an indeterminate number of Palm Beach County voters into voting for Buchanan rather than Gore as sufficient cause for a new election limited to that county, I suggested the possibility of such relief in Tribe, supra note 10. In that brief article, written just before I was asked to represent Gore but published the day after I undertook that representation (a fact unfortunately noted only in some editions of the New York Times), I made what in hindsight I have come to regard as the mistake of suggesting that a special election might properly be held after November 7 in a county (such as Palm Beach) where the ballot's design demonstrably led many voters to cast their ballots for the wrong candidate and where no amount of recounting could possibly cure the problem. Because Congress had by statute specified November 7, 2000 - the "Tuesday next after the first Monday in November," 3 U.S.C. 1 (1994) - as the day for "chusing the Electors," U.S. Const. art. II, 1, cl. 4, that federal statute would appear to forbid Florida to delay its choice pending a rerun of all or part of its election. An argument might be made, however, that because Article II, Section 1, Clause 4, requires national uniformity only with respect to "the Day on which [the Electors] shall give their Votes," id. (emphasis added) - a day that, under 3 U.S.C. 7, fell on December 18, 2000 - the attempt by Congress to set a uniform date for choosing electors - although otherwise within Congress's affirmative authority to "determine the Time of chusing the Electors," U.S. Const. art. II, 1, cl. 4 - would be trumped by Article II, Section 1, Clause 2's delegation to each state's "legislature" of the task of "directing" the "Manner" by which that "State shall appoint [its] Electors," U.S. Const. art. II, 1, cl. 2. On that view, whenever the state legislature's intent, as duly construed by the state's highest court, requires either recounting or recasting ballots in all or part of the state to ascertain what the true choice of the state's electorate was on Election Day, Congress is powerless to override that requirement - a requirement imposed, after all, in direct discharge of a duty delegated by Article II, Section 1, Clause 2 to the state legislature. On that reading, I was not mistaken in my original suggestion. However, the fact that Article II, Section 1, Clause 4 requires national uniformity only for the day when the electors are to vote does not mean that the provision of Article II, Section 2, Clause 4, empowering Congress to "determine the Time of chusing the Electors" precludes a congressional choice in favor of national uniformity for the day when the electors are to be chosen - a choice Congress plainly made in 3 U.S.C. 1.
The best way to harmonize the relevant provisions - although not a way easy to reconcile with the view of Article II, Section 1, Clause 2 taken in the concurring opinion of Chief Justice Rehnquist in Bush, 121 S. Ct. at 535 (Rehnquist, C.J., concurring) - accordingly seems to entail the view that state legislation enacted under Article II, Section 1, Clause 2 must conform to federal legislation (such as 3 U.S.C. 1) enacted pursuant to Article II, Section 1, Clause 4, even with respect to facets of such federal legislation not mandated by Article II, Section 1, Clause 4's limited national uniformity clause. One of the inescapable conclusions of that analysis is that I was indeed wrong in thinking, before being retained to represent the Vice President, that the Palm Beach County Election might lawfully be rerun in view of the misleading ballots it employed.
n134. Posner, supra note 56, at 108.
n135. Id.
n136. Id.
n137. Bush, 121 S. Ct. at 529.
n138. Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1220, 1237 n.53 (Fla. 2000) (quoting Boardman v. Esteva, 323 So. 2d 259, 269 (Fla. 1975)).
n139. Id. at 1237 n.52 (emphasis added) (quoting State ex rel. Landis v. Dyer, 148 So. 201, 203 (Fla. 1933)).
n140. Id. at 1237. To support this conclusion, the court referred in a footnote to a 1933 decision, State ex rel. Landis v. Dyer, 148 So. 201 (Fla. 1933), in which the Florida Supreme Court stated that "the Legislature may impose reasonable rules and regulations for ... governance [of the state constitutional right to vote], but it cannot under the guise of such regulation unduly subvest [sic] or restrain this right." Palm Beach, 772 So. 2d at 1237 n.52 (quoting Dyer, 148 So. at 203).
n141. Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring).
n142. Id.
n143. In addition, the Florida cases cited by the Chief Justice involve election contests in which the opposing party wanted to invalidate illegal votes. See id. at 536, 538. For example, both Boardman and Beckstrom v. Volusia County Canvassing Board, 707 So. 2d 720 (Fla. 1998), involved challenges to absentee ballots that had supposedly been cast contrary to statutory requirements. Thus, both contests were complaints about the "receipt of a number of illegal votes ... sufficient to change or place in doubt the result of the election," rather than claims concerning the "rejection of a number of legal votes." Fla. Stat. ch. 102.168(3)(c) (2000). In those cases, the court was concerned about what might constitute an "illegal vote," and had no need to focus on what might be a "legal vote."
n144. 772 So. 2d 1243 (Fla. 2000).
n145. Id. at 1255.
n146. See id.
n147. 399 So. 2d 1021 (Fla. Dist. Ct. App. 1981).
n148. Id. at 1022. On appeal, the losing candidate argued that the trial court should at least have held an evidentiary hearing on the validity of the ballots, rather than ruling on their status based on the information in the complaint. The appeals court denied relief, observing that the judgments made by county canvassing boards regarding the validity of votes were ""presumptively correct and if rational and not clearly outside legal requirements should be upheld ... .'" Id. at 1022-23 (quoting Boardman, 323 So. 2d at 268 n.5).
n149. See, e.g., Bolden v. Potter, 452 So. 2d 564, 567 (Fla. 1984) ("To require evidence which would establish with mathematical certainty the specific number of invalid votes sufficient to change the result of an election would make the task of setting aside an election because of intentional fraud or corruption virtually impossible ... .").
n150. Gore, 772 So. 2d at 1255 (quoting Fla. Stat. ch. 102.168(3)(c) (2000)).
n151. Posner, supra note 56, at 119.
n152. Id.
n153. Fla. Stat. ch. 102.141(4) (2000).
n154. Id. ch. 102.166(4)(a).
n155. Id. ch. 102.166(5).
n156. Id. ch. 101.5614(5).
n157. Posner, supra note 56, at 126 (emphasis added).
n158. Florida circuit courts have statewide jurisdiction. See Fla. Const. art. 5, 5; Fla. Stat. ch. 26.012 (2000). The Florida Supreme Court ordered the circuit court in Leon County to oversee the recount. Gore v. Harris, 772 So. 2d 1243, 1262 (Fla. 2000) (ordering Florida's circuit court in Leon County "to immediately tabulate by hand [approximately] 9000 Miami-Dade ballots," and granting it the authority "to enter such orders as are necessary to add any legal votes to the total statewide certifications and to enter any orders necessary to ensure the inclusion of the additional legal votes for Gore in Palm Beach County and the 168 additional legal votes from Miami-Dade County" (footnote omitted)); id. ("Moreover, since time is also of the essence in any statewide relief that the circuit court must consider, any further statewide relief should also be ordered forthwith and simultaneously with the manual tabulation of the Miami-Dade undervotes.").
n159. See Fla. Stat. ch. 101.5614(4) (2000) (describing the procedure for manually counting write-in ballots); id. ch. 101.5614(5) (detailing the procedure for tabulating "damaged or defective" ballots); id. ch. 102.166(5) (authorizing canvassing boards to decide to "manually recount all ballots" when a preliminary manual recount "indicates an error in vote tabulation").
n160. Fla. Stat. ch. 102.166(7)(a) (2000) ("A counting team must have, when possible, members of at least two political parties.").
n161. Id. ch. 102.166(6) ("Any manual recount shall be open to the public.").
n162. See Opening Brief for Appellants at 7, Siegel v. LePore, 234 F.3d 1163 (11th Cir. 2000) (No. 00-15981) (claiming that, during the protest phase of the election, several of the counting teams in Palm Beach County did not include Republicans); id. (alleging that the canvassing board members in Palm Beach County asked only Democrats to monitor the recount process).
n163. See Fla. Stat. ch. 101.5614(8) (2000) ("The return printed by the automatic tabulating equipment, to which has been added the return of write-in, absentee, and manually counted votes, shall constitute the official return of the election.").
n164. See infra note 376.
n165. See Barstow & Van Natta, supra note 114.
n166. Cf. Siegel v. LePore, 120 F. Supp. 2d 1041, 1051 n.10 (S.D. Fla. 2000) ("It should be noted that any presidential candidate was afforded an equal opportunity under the statute to ask for a manual recount in each Florida county.").
n167. The Florida election code requires that a recount team "must have, when possible, members of at least two political parties," Fla. Stat. ch. 102.166(7)(a) (2000), and makes no distinction between major and minor parties. Thus, it is possible that members of minor parties could have participated in the monitoring process.
n168. I say "ostensibly" because behind each machine, and often at its controls, there sits a human designer or programmer.
n169. See The Federalist No. 51, at 331 (James Madison) (Robert Scigliano ed., 2000).
n170. Cf. Siegel v. LePore, 234 F.3d 1163, 1211-13 (11th Cir. 2000) (Carnes, J., dissenting) (arguing that the Florida election code on its face violates the Equal Protection Clause because it permits a candidate or political party to request recounts only in populous counties that tend to support that candidate, thereby diluting the votes of citizens in less populous regions).
n171. Boardman v. Esteva, 323 So. 2d 259, 268 (Fla. 1975).
n172. See Fla. Stat. ch. 102.166(6) (2000) ("Any manual recount shall be open to the public.").
n173. See Bush, 121 S. Ct. at 531-32; see, e.g., Posner, supra note 56, at 122. ("The court had given Gore the votes he had received in Miami-Dade's partial recount even though the full recount might never be completed ... .").
n174. Boardman, 323 So. 2d. at 263.
n175. Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring).
n176. See, e.g., Posner, supra note 66, at 26 (describing Article II as a tool to be utilized by the Supreme Court to "prevent[] the state courts from hijacking an election by changing the rules after the outcome of the election is known").
n177. 475 U.S. 260 (1986).
n178. Id. at 270.
n179. 520 U.S. 351 (1997).
n180. Id. at 369-70.
n181. Justice Souter dissented in Timmons, but on narrow procedural grounds. See id. at 382-84 (Souter, J., dissenting).
n182. This trend has been noted elsewhere. See, e.g., Pildes, supra note 67, at 704.
n183. For an insightful analysis of the Justices' inclinations to favor either rules or standards, and the implications of those tendencies, see generally Kathleen M. Sullivan, The Supreme Court, 1991 Term - Foreword: The Justices of Rules and Standards, 106 Harv. L. Rev. 22 (1992).
n184. Cf. John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (1980) (arguing that the Supreme Court should intervene primarily to reinforce the efficacy of political representation and participation, not to undermine or circumvent them).
n185. See infra pp. 257-58.
n186. Although it is possible to assign machine-rejected punchcard ballots with reasonable accuracy into no more than five or six categories (chads perforated on three corners, on two corners, on one corner, not perforated at all but substantially dimpled, only mildly dimpled), ballots on which the voter writes something can differ along innumerable dimensions, and on each dimension or axis the range of possibilities varies continuously and cannot usefully be subdivided into some finite number of categories or types. Put simply, there are literally an infinite number of ways to write "Al Gore" or "George Bush."
n187. See infra pp. 287-88.
n188. See, e.g., Op-ed, A Gore Coup d'Etat?, Wall St. J., Nov. 10, 2000, at A18; see also Noemie Emery, First Principles in Florida: Conservatives Believe in Rules, Liberals Want To Be "Fair," Weekly Standard, Dec. 11, 2000, at 24-25 ("The Democrats are the party of the dimpled chad, from which one tries to intuit an unknown voter's intention.").
n189. The fact that there was a magistrate with statewide jurisdiction who could in theory have replaced the legislature's "intent of the voter" standard with something more hard-edged and objective-sounding tells us absolutely nothing about whether equal protection or due process norms mandated such a replacement. See infra section A.3(b), pp. 237-47.
n190. See Bush, 121 S. Ct. at 530-31.
n191. This statement in some respects echoes the Supreme Court's observation that "the Fourth Amendment protects people, not places." Katz v. United States, 389 U.S. 347, 351 (1967).
n192. Bush, 121 S. Ct. at 529, 532.
n193. Peter Shaffer, Amadeus act 1, sc. 8, at 29 (1st U.S. ed. 1981).
n194. Bush, 121 S. Ct. at 530.
n195. Some commentators assume as much and treat Bush v. Gore as an innovative extension of equal protection law into the realm of vote-counting. See Jack M. Balkin, Bush v. Gore and the Boundary Between Law and Politics, 110 Yale L.J. 1407, 1426 (2001) ("The majority's equal protection argument is novel, but not entirely crazy. It extends the principle of "one person, one vote' from the question of how districts are apportioned before the election to the question of how the votes are tabulated after the election."). I confess that I fail to see why vote-counting should be regarded as a new arena for the application of equal protection analysis. Even under traditional doctrines, counting the ballots of Palm Beach County twice and those of Broward County just once, for instance - giving Palm Beach voters twice as much influence on the outcome as Broward voters - would violate the Equal Protection Clause. What was new was what the Court did with equal protection analysis, not what it did it to.
n196. 377 U.S. 533 (1964).
n197. Cf. id. at 562-63.
n198. 146 U.S. 1 (1892).
n199. Id. at 10.
n200. Bush, 121 S. Ct. at 530 (citing Harper v. Virginia Board of Elections, 383 U.S. 663, 665 (1966)).
n201. See Hunt v. Cromartie, 121 S. Ct. 1452, 1458 (2001) (finding no equal protection violation "where the State has articulated a legitimate political explanation for its districting decision, and the voting population is one in which race and political affiliation are highly correlated").
n202. 478 U.S. 109 (1986).
n203. In congressional elections, the permitted deviation between the size of the most and least populous districts is very small. See White v. Weiser, 412 U.S. 783, 790-93 (1973). However, in other elections the permitted deviance is much higher. See Brown v. Thomson, 462 U.S. 835, 842-43 (1983).
n204. In Gaffney v. Cummings, 412 U.S. 735 (1973), the Court held that voting districts may be apportioned in accordance with expected voting patterns:
Politics and political considerations are inseparable from districting and apportionment. The political profile of a State, its party registration, and voting records are available precinct by precinct, ward by ward. These subdivisions may not be identical with census tracts, but, when overlaid on a census map, it requires no special genius to recognize the political consequences of drawing a district line along one street rather than another. It is not only obvious, but absolutely unavoidable, that the location and shape of districts may well determine the political complexion of the area. District lines are rarely neutral phenomena. They can well determine what district will be predominantly Democratic or predominantly Republican, or make a close race likely. Redistricting may pit incumbents against one another or make very difficult the election of the most experienced legislator. The reality is that districting inevitably has and is intended to have substantial political consequences.
Id. at 753.
n205. The expected error rate could even be determined by the state, see McPherson v. Blacker, 146 U.S. 1, 10 (1892), provided that the rate stayed within a margin of about ten percent. See Brown, 462 U.S. at 842 ("Minor deviations from mathematical equality among state legislative districts are insufficient to make out a prima facie case of invidious discrimination under the Fourteenth Amendment so as to require justification by the State ... [and] an apportionment plan with a maximum population deviation under 10% falls within this category of minor deviations." (citation omitted)).
n206. One further point: Reynolds and Davis do not require the parties to be ignorant ex ante of the likely outcome of apportionment for it to be deemed fair. On the contrary, apportionment is invariably designed in full knowledge of its probable consequences. See Davis, 478 U.S. at 128-29 (1986). Under this sort of equal protection claim, it does not matter if the vote-counters have seen the exit polls and know who is winning and by how much, or are awaiting absentee ballots and know what difference they will make, or otherwise understand the consequences of the count for a particular candidate. So long as the counting process is fair and is not deliberately designed to miscount votes, it will withstand Reynolds-style scrutiny.
n207. 372 U.S. 368 (1963).
n208. Bush, 121 S. Ct. at 531 (citing Gray, 372 U.S. 368).
n209. 394 U.S. 814 (1969).
n210. Bush, 121 S. Ct. at 531 (quoting Moore, 394 U.S. at 819) (internal quotation marks omitted).
n211. Moore, 394 U.S. at 818.
n212. Id. at 819.
n213. Indeed, the Court's own choice of terms - "equal treatment" rather than "equal protection" - may indicate an awareness that although voters in some sense might have been treated unequally, none was denied equal protection of the laws of Florida. Village of Willowbrook v. Olech, 120 S. Ct. 1073, 1073 (2000), is not to the contrary. The Court there held only that a complaint alleging state action intentionally singling the plaintiff out for comparatively adverse treatment, either for no reason at all or in a "spiteful effort to "get' him for reasons wholly unrelated to any legitimate state objective[,] ... gives rise to a cause of action on behalf of a "class of one.'" Id. at 1074 (quoting Olech v. Village of Willowbrook, 160 F.3d 386, 387 (7th Cir. 1998) (quoting Esmail v. Macrane, 53 F.3d 176, 180 (7th Cir. 1995))) (internal quotation marks omitted).
n214. Id.
n215. 509 U.S. 630 (1993) (holding that plaintiffs could withstand a motion to dismiss because they had made a sufficient showing that race may have been the predominant consideration of the state legislature in drawing a congressional district). The Shaw progeny include Hunt v. Cromartie, 121 S. Ct. 1452 (2001); Sinkfield v. Kelley, 121 S. Ct. 446 (2000); Bush v. Vera, 517 U.S. 952 (1996); Shaw v. Hunt, 517 U.S. 899 (1996); Miller v. Johnson, 515 U.S. 900 (1995); and United States v. Hays, 515 U.S. 737 (1995).
n216. This latter inquiry can be regarded as a specific form of vote dilution.
n217. Professor Karlan thinks that these cases fail to recognize that "the very purpose of apportionment is to aggregate voters into groups for the purpose of electing representatives." Pamela S. Karlan & Daryl J. Levinson, Why Voting Is Different, 84 Cal. L. Rev. 1201, 1204 (1996); id. ("The "concept of "representation" necessarily applies to groups: groups of voters elect representatives, individual voters do not.'" (quoting Davis v. Bandemer, 478 U.S. 109, 167 (Powell, J., concurring in part and dissenting in part))).
n218. There is a virtual ban on looking at surrounding districts. See Sinkfield, 121 S. Ct. at 447 (limiting standing to bring Shaw claims to residents of the challenged district); Hays, 515 U.S. at 747. However, only Shaw claims are limited to the voters in a particular district; other types of voting injury claims require a comparison of districts in order to determine whether there has been an injury. Hays responds to the unique problems posed for standing under Shaw: district lines are not drawn in isolation, and the border of one district is simultaneously the border of the district adjacent to it. If, for example, a majority-minority district is surrounded by majority-white districts, then voters in the adjoining districts could be characterized as "apportioned out" on the basis of race. Because the manner in which one district is apportioned or reapportioned affects and is affected by its neighboring districts, race-based apportionment can "ripple" out into the surrounding districts, leaving each of them subject to challenge under Shaw. This "ripple effect" was discussed in oral argument before the Supreme Court. Transcript of Oral Argument, Shaw (No. 94-923), 1995 WL 729891, at 7-8 (Dec. 5, 1995).
n219. "The practical consequences of the Court's choice of the atomized individual as the building block for legislative districting are profound. First, it has essentially judicialized the process of reapportionment. . . . Second . . . [in the Shaw cases, it] has turned increasingly toward a liability standard that focuses on the individual." Pamela S. Karlan, Politics by Other Means, 85 Va. L. Rev. 1697, 1719-20 (1999). At one point, Professor Karlan traces the rise of this individualism to Reynolds. Id. at 1716. However, although Reynolds, after a fashion, emphasizes the rights of individuals, the thrust of Reynolds (as Karlan acknowledges elsewhere, see Samuel Issacharoff & Pamela S. Karlan, Standing and Misunderstanding in Voting Rights Law, 111 Harv. L. Rev. 2276, 2292 (1998)), is to assert the individual's rights through her membership in a group. One-person, one-vote challenges, although concerned with the relative weight of individual votes, consider those votes in the aggregate. To ensure that each vote was counted equally, the Reynolds Court looked to the difference in numbers of voters between the plan's most populous and least populous districts. Thus, the equality proposed by Reynolds was the equality of groups of voters as split into voting districts. The Court's use of Reynolds in Bush v. Gore, 121 S. Ct. at 530, is not an extension of the one-person, one-vote doctrine, but a bastardization of it. It was, however, a bastardization that is precisely in line with the Court's Shaw jurisprudence.
n220. Shaw, 509 U.S. at 649.
n221. Or perhaps more accurately, by having his race used as a proxy for predicted voting behavior. See id. at 642.
n222. Whether, that is, he suffers anything more than race-based stigma.
n223. Pamela S. Karlan, The Newest Equal Protection: Regressive Doctrine on a Changeable Court, in The Vote, supra note 3, at 77, 78.
n224. Professor Karlan argues that the Court has used this new model both to extend its reach over political controversies by "essentially judicializing the process of reapportionment," Karlan, supra note 219, at 1719-20, and at the same time to provide a "framework that ignores politics' role as "a device for regulating, rationing, and apportioning political power among ... groups.'" Pamela S. Karlan, Just Politics? Five Not So Easy Pieces of the 1995 Term, 34 Hous. L. Rev. 289, 293 (1997) (quoting Holder v. Hall, 512 U.S. 874, 893 (1994) (Thomas, J., concurring in the judgment) (criticizing the Supreme Court's Voting Rights Act jurisprudence for treating the Act as such a device)). This model in turn frames the decision in Bush v. Gore, according to Karlan: "Whatever interest the Supreme Court's decision vindicated, it was not the interest of an identifiable individual voter. Rather, it was a perceived systemic interest in having recounts conducted according to a uniform standard or not at all. It was structural equal protection, just as the Shaw cases have been." Karlan, supra note 223, at 89-90.
n225. For example, the Court's recent federalism jurisprudence has required Congress to make detailed factual findings to justify any remedy proposed under its Fourteenth Amendment power to abrogate state sovereign immunity, and it has forced Congress to limit its remedies to those that the Court, on its de novo review of the record, deems "congruent and proportional to the targeted violation ... [lest] Congress ... rewrite the Fourteenth Amendment law laid down by this Court." Bd. of Trs. of Univ. of Ala. v. Garrett, 121 S. Ct. 955, 967-68 (2001).
n226. See Issacharoff & Karlan, supra note 219; Pamela S. Karlan, The Fire Next Time: Reapportionment After the 2000 Census, 50 Stan. L. Rev. 731 (1998).
n227. The Justices who dissented in Shaw have elsewhere noted the strange nature of the injury described in that case. See Bush v. Vera, 517 U.S. 952, 1005 (1996) (Stevens, J., dissenting) ("The Court has, with its "analytically distinct' jurisprudence of racial gerrymandering, struck out into a jurisprudential wilderness that lacks a definable constitutional core and threatens to create harms more significant than any suffered by the individual plaintiffs challenging these districts." (citation omitted)); id. at 1045 (Souter, J., dissenting) ("When the Court devises a new cause of action to enforce a constitutional provision, it ought to identify an injury distinguishable from the consequences of concededly constitutional conduct, and it should describe the elements necessary and sufficient to make out such a claim... . Although the Court has repeatedly disclaimed any intent to go as far as to outlaw all conscious consideration of race in districting, after three rounds of appellate litigation seeking to describe the elements and define the contours of the Shaw cause of action, a helpful statement of a Shaw claim still eludes this Court."). There has also been a wealth of scholarly commentary on the type of standing created by Shaw. See, e.g., T. Alexander Aleinikoff & Samuel Issacharoff, Race and Redistricting: Drawing Constitutional Lines After Shaw v. Reno, 92 Mich. L. Rev. 588 (1993); John Hart Ely, Gerrymanders: The Good, the Bad, and the Ugly, 50 Stan. L. Rev. 607, 628 (1998) [hereinafter Ely, The Good, the Bad, and the Ugly]; John Hart Ely, Standing To Challenge Pro-Minority Gerrymanders, 111 Harv. L. Rev. 576 (1997) [hereinafter Ely, Standing]; Samuel Issacharoff & Thomas C. Goldstein, Identifying the Harm in Racial Gerrymandering Claims, 1 Mich. J. Race & L. 47, 51-58 (1996); Issacharoff & Karlan, supra note 219; Laughlin McDonald, Can Minority Voting Rights Survive Miller v. Johnson?, 1 Mich. J. Race & L. 119, 141-44 (1996).
n228. Ely, Standing, supra note 227, at 584.
n229. See id.
n230. See id. at 585 (noting that "filler people" are singled out for disadvantage in Shaw-style districting). Professors Issacharoff and Karlan object to Professor Ely's analysis for two reasons that seem to me immaterial here. First, they note that the injury is not confined to filler people but extends to every member of a challenged district: minority voters as well as filler people have standing to bring a Shaw claim. See Issacharoff & Karlan, supra note 219, at 2278 (noting that the Supreme Court "has never required that the plaintiffs in Shaw cases be white"). The race of the injured voter is unimportant because the injury results from the stigma that all voters in a given district suffer from apportionment on the basis of race. See Miller v. Johnson, 515 U.S. 900, 911 (1995) ("The essence of ... [a] Shaw [claim] is that the State has used race as a basis for separating voters into districts. Just as the State may not, absent extraordinary justification, segregate citizens on the basis of race in its public parks, buses, golf courses, beaches, and schools, so did we recognize in Shaw that it may not separate its citizens into different voting districts on the basis of race." (citations omitted)); see also id. at 915 (acknowledging injury when the state assigns voters to districts on the basis of race); Shaw v. Reno, 509 U.S. 630, 645 (1983) (same). That, however, is precisely Ely's point: at least some of the plaintiffs do have standing as it is traditionally understood. Issacharoff and Karlan also object that the injury Ely attributes to filler people - traditional vote-dilution - is not the same as the injury established in Shaw - that of being assigned to a district on the basis of race. See Issacharoff & Karlan, supra note 219, at 2285-87. Indeed, the Shaw plaintiffs chose to assert an injury quite different from vote dilution. See id. at 2284 ("The Shaw plaintiffs deliberately declined to claim racial vote dilution, and thus their standing cannot rest on the injury of having "less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice.'" (quoting White v. Regester, 412 U.S. 755, 766 (1973))). Ely's point, however, is that the Shaw plaintiffs, or at least a subset of them, could have established standing under traditional vote-dilution analysis. He seems plainly right.
n231. Bush certainly had standing to argue on Article II grounds that the recount was unconstitutional. That he had standing under Article II for that purpose does not, of course, imply that he had such standing to press the equal protection claims of voters. See Miller v. Albright, 523 U.S. 420, 451 (1998) (O'Connor, J., concurring in the judgment) (discussing concerns in allowing third parties to press constitutional claims on behalf of others); U.S. Dep't of Labor v. Triplett, 494 U.S. 715, 720 (1990) (describing the constitutional requirements for third-party standing).
n232. See Karlan, supra note 223, at 85; cf. United States v. Hays, 515 U.S. 737, 747 (1994). I am not persuaded by Professor Chemerinsky's suggestion that Hays provides a basis for denying then-Governor Bush standing to bring his equal protection challenge. See Erwin Chemerinsky, Bush v. Gore Was Not Justiciable, 76 Notre Dame L. Rev. 1093, 1098 (2001). Hays was about apportionment; Bush v. Gore was about counting votes. There is no requirement to adopt Hays as a paradigm for all equal protection voting rights cases. For example, to establish standing under Reynolds, a plaintiff need not reside in the challenged district, but only in the particular state. Reynolds v. Sims, 377 U.S. 533, 563 (1964). (I am, however, in full agreement with Professor Chemerinsky's conclusion that Bush v. Gore was not justiciable. See infra Part V, pp. 275-92.)
n233. Bush presented his claim as an action on behalf of Florida voters. He argued that the Equal Protection Clause "forbids the state from treating similarly situated voters differently based merely on where they live." See Brief for Petitioners, supra note 97, at 18. For the reasons set forth in the text that follows, I disagree with Professor Karlan's contention that "unless and until the Supreme Court is prepared to say that [Bush] supporters are disproportionately likely not to have their votes recovered under the prescribed process - and the Court made no such finding - he is an especially unlikely candidate for third-party standing." Karlan, supra note 223, at 85.
n234. Acknowledging this injury, Professor Karlan argues in effect that the Court was hoist by its own political petard: either the Supreme Court had to find that Bush voters were disproportionately affected, or the Court would have to characterize Bush as an appropriate representative for the undervoters in Broward County - which, she claims, is a rather odd position for the Court to take, inasmuch as Bush was attempting to have their votes excluded rather than included. Thus, according to Karlan, it is precisely the Supreme Court's refusal to recognize the partisan nature of Bush's claim that deprives him of third-party standing, while at the same time enabling the Court to take the case on equal protection grounds. See Karlan, supra note 223, at 85-86. But Karlan construes Bush's third-party standing too narrowly. Bush could claim standing to represent all voters whose votes were miscounted, even though his ultimate interest was in ensuring that some of their votes not be tallied, just as the white defendant in Powers v. Ohio, 499 U.S. 400 (1991), could claim to represent all prospective jurors excluded from a jury pool because of their race, including the black prospective jurors, even though his interest would be to exclude only those he felt were likely to convict him. See id. at 415.
n235. Indeed, Professor Karlan seems to concede as much. See Karlan, supra note 223, at 85-86.
n236. 499 U.S. 400 (1991).
n237. See id. at 413-14; 1 Tribe, supra note 123, 3-19, at 441-42. In fact, the Shaw Court cited Powers to establish that racial classifications "receive close scrutiny even when they ... burden or benefit the races equally." Shaw v. Reno, 509 U.S. 630, 651 (1993).
n238. For example, a voter typically cannot know whether his or her vote has been miscounted. Cf. Powers, 499 U.S. at 414-15 (noting similarly significant obstacles any individual juror would face in asserting an equal protection claim).
n239. Powers is instructive here. The defendant in that case had an interest in ensuring a fair process. Thus, he may have ensured that some jurors whom he would later strike using his peremptory challenges would be included in the jury pool. Similarly, Bush's professed interest was in a fair process; the fact that some votes would be excluded was not the point. The rather surreal argument that suggests Bush did not belong in Bush v. Gore is, frankly, what gives lawyers a bad name. It's the constitutional equivalent of asking what the definition of "is" is. Cf. Terence Hunt, A Different Clinton Appears on Tape, Associated Press Online, Sept. 17, 1998, available at 1998 WL 6724483.
n240. The structural equal protection idea seems to echo in some respects a notion I first advanced some time ago. See Laurence H. Tribe, Structural Due Process, 10 Harv. C.R.-C.L. L. Rev. 269, 300-01 (1975).
n241. Id.
n242. 391 U.S. 510 (1968).
n243. Id. at 519-20. There are a variety of ways by which the Court may bring constitutional principles to bear on decisionmaking arrangements to create linkages between those arrangements and changing technology or changing values or both. Witherspoon illustrates forging that linkage by forcibly opening the relevant decisionmaking tribunal - there, the jury - to participation by persons with views that challenge the current orthodoxy. In other cases, the Court achieves linkage by defining rights or responsibilities, particularly at the boundaries where they interface with competing interests, in terms that build in one or more key variables drawn from adjacent fields of endeavor. Elsewhere I have examined the Court's definition of fetal viability in Roe v. Wade, 410 U.S. 113 (1973), from this perspective. In Structural Due Process, Tribe, supra note 240, I describe how defining viability in terms of the ability to survive outside the womb enables a polity to invest public revenues in medical research that brings closer to conception the point after which abortion may be banned except when needed for the life or health of the mother. By linking such authority to public investment, Roe v. Wade ensures that those committed to preventing avoidable fetal destruction bear more of the costs, rather than imposing those costs solely on the women involved. See id. at 297-98.
n244. Witherspoon, 391 U.S. at 519 n.15 (quoting Trop v. Dulles, 356 U.S. 86, 101 (1958)).
n245. 426 U.S. 88 (1976).
n246. Id. at 114.
n247. Id. ("Th[e Civil Service Commission] has no responsibility for foreign affairs, for treaty negotiations, for establishing immigration quotas or conditions of entry, or for naturalization policies. Indeed, it is not even within the responsibility of the Commission to be concerned with the economic consequences of permitting or prohibiting the participation by aliens in employment opportunities in different parts of the national market. On the contrary, the Commission performs a limited and specific function.").
n248. Id. at 115. Professor Elena Kagan has applied a structural due process argument to suggest that, if the decision in Hampton were made by the President rather than by an administrative agency, the resulting political accountability could render the decision constitutional. See Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2371-72 (2001) (citing 1 Tribe, supra note 123, 17-2, at 1682).
n249. 438 U.S. 265 (1978).
n250. Id. at 309.
n251. 121 S. Ct. 1281 (2001).
n252. Id. at 1291-92.
n253. Id. at 1292 (emphasis omitted); see also id. at 1294 (Kennedy, J., concurring) (stressing the police department's role in formulating the policy). In Ferguson, as in Bush, Justice Kennedy went out of his way to describe the controversy as one the Court was compelled to "take ... as it comes to us." Id. at 1295.
n254. "The Supreme Court as an institution seems increasingly confused, or indifferent, about what politics is for and when courts need to regulate the process ... . Having embarked on the course of resolving political questions into judicial ones, it is notable how many of these judicial questions the Court fails to resolve, or to resolve in a doctrinally coherent and stable way." Karlan, supra note 224, at 290-92.
n255. I offer no view here of the explanatory power of "structural equal protection" in other contexts or for other cases.
n256. Bush, 121 S. Ct. at 532 ("It is obvious that the recount cannot be conducted in compliance with the requirements of equal protection and due process without substantial additional work. It would require not only the adoption (after opportunity for argument) of adequate statewide standards for determining what is a legal vote, and practicable procedures to implement them, but also orderly judicial review of any disputed matters that might arise."). The Court made only one other explicit reference to the alleged due process violations in the Florida recount. See id. at 529 (declaring the questions presented to include potential violations of both the Equal Protection and Due Process Clauses).
n257. See, e.g., Bishop v. Wood, 426 U.S. 341, 344-45 (1976). Chief Justice Rehnquist would have gone further still. See Cleveland Bd. of Educ. v. Loudermill, 470 U.S. 532, 559 (1985) (Rehnquist, J., dissenting).
n258. Bush, 121 S. Ct. at 530.
n259. See Bd. of Regents of State Colls. v. Roth, 408 U.S. 564, 569 (1972).
n260. Overvotes are ballots that are rejected because they appear to contain more than one vote for a given office.
n261. Bush, 121 S. Ct. at 531.
n262. Some scholars have echoed this argument. Most prominently in the defenders' camp has been Professor Nelson Lund. Lund argues:
In this case, the Florida court devised an extremely complex system of weighting, in which certain kinds of ballots were more likely to be counted as legal votes in some places than in others, thus discriminating for and against different groups of voters based on where they happened to reside. Most obviously, voters who cast "overvote" ballots in Broward, Palm Beach, and Miami-Dade Counties were treated more favorably than those who cast similar ballots elsewhere. Similarly, voters who cast "dimpled chad" ballots in Broward were treated more favorably than those who cast similar ballots in Palm Beach. Voters living in the unrecounted (and more Republican) precincts of Miami-Dade were disadvantaged in comparison with those living in the recounted (and more Democratic) precincts. The complexity of the vote dilution involved did not convert it into something other than vote dilution.
Nelson Lund, The Unbearable Rightness of Bush v. Gore 30-31 (May 7, 2001) (unpublished manuscript) (citations omitted), available at http://www.gmu.edu/departments/law/faculty/papers/ docs/01-17.pdf. But Lund's assumption that counting overvotes from a given county means treating its voters "more favorably" is simply false. See infra pp. 257-58. Lund then imagines that the Florida court's justification for these selective recounts was that any "legal votes" that happened to turn up should have been added to the totals generated by the machine counts. Lund, supra, at 34. Lund then proceeds to mock his own caricature of the state court's standard:
Suppose, for example, that "undervote" ballots containing evidence of an intent to vote for Gore were changed to "legal votes," but similar ballots showing an intent to vote for Bush were not changed to legal votes. Such a recount would be "better" than the machine counts under the criterion employed by the Florida court because it would result in more "legal votes" being tabulated. But it would not be better in any constitutionally relevant sense, or indeed under any sane criterion. The difference between this hypothetical and the actual order of the Florida court is only one of degree, and a very slight degree at that.
Id. at 34-35 (citations omitted). But, of course, the difference is one of kind, not degree.
n263. Florida Supreme Court Chief Justice Wells made a similar point in his dissent from that court's December 8 decision. See Gore v. Harris, 772 So. 2d 1243, 1264 n.26 (Fla. 2000) (Wells, C.J., dissenting).
n264. Bush, 121 S. Ct. at 531.
n265. See McConnell, supra note 78, at 658 n.8 (citing Gore v. Harris, 773 So. 2d 524, 535 n.26 (Fla. 2000) (Pariente, J., concurring)); Mintz & Slevin, supra note 102, at A20 (noting one supervisor's comment that "if it's the same person, it's a no-brainer" to explain why overvotes were included in the tally despite the technical flaw of marking "two" candidates, and noting that thirty-four counties included such "overvotes").
n266. The Court's insistence that there be a uniform standard more specific than the "clear intent of the voter" - and its holding that, so far as equal protection is concerned, any standard sufficiently objective and uniform would do - assumes that whole classes of certain "ballot markings" may properly be excluded. One must wonder how the Court could require a uniform standard that would undoubtedly exclude a "class" of ballots (say, dimpled chads) and yet find a constitutional violation in the state legislature's decision to exclude a different class of ballots (that is, overvotes).
n267. See Fla. Stat. ch. 101.5614(6) (2000) ("If an elector marks more names than there are persons to be elected to an office or if it is impossible to determine the elector's choice, the elector's ballot shall not be counted for that office, but the ballot shall not be invalidated as to those names which are properly marked.").
n268. Bush, 121 S. Ct. at 530.
n269. See, e.g., County of Sacramento v. Lewis, 523 U.S. 833, 847 n.8 (1998); Washington v. Glucksberg, 521 U.S. 702, 703 (1997). See generally 1 Tribe, supra note 123, 7-6, at 1325-26, 8-2, at 1335-43, 8-8, at 1362-65, 1367-71.
n270. See, e.g., Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833, 851 (1992).
n271. See, e.g., E. Enters. v. Apfel, 524 U.S. 498, 550 (1998) (Kennedy, J., concurring) (finding a multi-million dollar exaction violative of substantive due process because it was qualitatively unforeseeable and quantitatively disproportionate); see also id. at 556-59 (Breyer, J., dissenting) (applying a form of substantive due process virtually indistinguishable from Justice Kennedy's but reaching the opposite conclusion).
n272. See, e.g., County of Sacramento, 523 U.S. at 861 (Scalia, J., concurring in the judgment); see also Saenz v. Roe, 526 U.S. 489, 528 (1999) (Thomas, J., dissenting); Cruzan v. Dir., Mo. Dep't of Health, 497 U.S. 261, 293-94 (1990) (Scalia, J., concurring).
n273. Collins v. City of Harker Heights, 503 U.S. 115, 125 (1992) (unanimous).
n274. See 1 Tribe, supra note 123, 7-5, at 1317-20, 8-1, at 1332-34.
n275. See, e.g., Reynolds v. Sims, 377 U.S. 533, 561-62 (1964).
n276. See Harper v. Va. Bd. of Elections, 383 U.S. 663 (1966) (invalidating a Virginia statute that conditioned otherwise qualified residents' right to vote on payment of a poll tax).
n277. Reynolds, 377 U.S. at 562.
n278. Some commentators and advocates disagree, buying into a superficial, appearance-driven idea of equality. Some even seem to see it as the only constitutional alternative. See, e.g., Lund, supra note 262, at 37. In his brief for the Florida Legislature, for example, Professor Charles Fried suggested that the "intent of the voter" standard is "capacious and unconstraining," Brief of the Florida House of Representatives and Florida Senate as Amici Curiae in Support of Neither Party and Seeking Reversal at 25, Bush, 121 S. Ct. 525 (2000) (No. 00-949), available at 2000 WL 1868121, at 25, and so "vague and devoid of guidance as to invite arbitrary or discriminatory treatment." Id. at 26. Professor Fried concluded that "there must be "narrow, objective and definite standards' governing the exercise of official discretion," id. at 27 (quoting Forsyth County v. Nationalist Movement, 505 U.S. 123, 131 (1992)), and that perhaps state election codes must, as a federal constitutional matter, reject votes not cast in accord with state-enumerated procedures that are capable of mechanical, nondiscretionary application, id.
Although appealing in its elegant simplicity, that argument reifies the ballot without much consideration of what it means to vote. Professor Fried places all the weight on the means of voting, and in so doing may be losing sight of the end: an election is not a test of the electorate's ability to follow directions, but rather a process whose goal is to aggregate the preferences of voters. Indeed, the point of voting, and the only reason to follow instructions, is to reveal the voter's intent, even if one's intent is "revealed" through the secret ballot only to an official aggregating device rather than publicly and verifiably identified as one's individual view. (It is largely because of this specialized, official, and aggregative character of the vote that voting has not been protected in the same manner, or to the same degree, as the unofficial speech of private individuals. See, e.g., Burdick v. Takushi, 504 U.S. 428, 441 (1992) (holding that a Hawaii prohibition on write-in voting does not violate its citizens' First Amendment right of free speech).) As such, the only justifiable basis for excluding a vote cast by an eligible voter who has made an apparently good-faith effort to comply with reasonable voting procedures is that the voter's intent cannot be determined, or that efforts to determine it would prove impracticable. Indented squares and hanging chads in many cases can and will clearly reveal the voter's intent in the context of the entire ballot against the backdrop of the particular polling place, its equipment, and its instructions. Ballots that do reveal the voter's intent should not be excluded solely because the voter behind the ballot did not succeed in following the instructions fully.
Moreover, Professor Fried's argument assumes that only votes cast in accordance with objectively administrable rules can reliably be treated as revealing intent. But a little reflection on recounts in general suggests that this is simply not the case. In fact, without a considerable degree of discretion, election officials would be unable to tailor the recount criteria to fit the unique circumstances of their particular counties; they would similarly be unable to catch the many different ways in which a voter can fail to follow directions and yet still unequivocally reveal her intended meaning. For example, Professor Fried's argument suggests that a ballot with only a dimpled chad and the following note should be rejected in a jurisdiction whose rules tell voters they must punch all the way through: "Vote tabulators, please take notice: I tried and tried to punch this damn thing all the way through in the "Bush' column for president, but no matter how often or how hard I tried, the stylus just wouldn't poke through. Some junk seemed to be stuck behind the correct space, and I just couldn't make a hole. But know this: I meant to, and attempted to, vote for George W. Bush for President." (I owe this example to a private communication from Mark Tushnet, June 2001.) Failing to count such a vote seems plainly wrong because the voter's intent is itself so clear. Although one could counter with a "positivist" due process argument (that there is no positive entitlement to have an incorrectly cast vote counted), see supra pp. 233-34 (discussing the lack of an entitlement), and with a rote equal protection answer (that if no ballots with writing were read, those ballots as a class would be treated uniformly across counties), the manifest arbitrariness of refusing to count so clear a ballot might make doing so a violation of due process and equal protection. One might argue that, if some voters' ballots in other counties are considered valid although they communicated with only a "dimple," surely this more explicit communication should suffice as well. There is, in the end, something otherworldly about the whole idea of squeezing all room for discretionary judgments from the vote-tallying process to banish every last opportunity for partisan fudging and cheating. At a minimum, that project would require the elimination of all manual counting, including manual processing of overseas and other absentee ballots. Whatever else may be said of such a venture, it surely cannot be required by a Constitution that functioned for the better part of two centuries on the basis of entirely manual ballot counting.
When Professor Fried served on the Supreme Judicial Court of Massachusetts, he appears to have recognized as much in an opinion he joined:
The cardinal rule for guidance of election officers and courts in cases of this nature is that if the intent of the voter can be determined with reasonable certainty from an inspection of the ballot, in the light of the generally known conditions attendant upon the election, effect must be given to that intent ... .
Delahunt v. Johnston, 671 N.E.2d 1241, 1243 (Mass. 1996) (emphasis added) (quoting O'Brian v. Election Commissioners, 153 N.E. 553 (1926)) (internal quotation marks omitted). To be fair, Professor Fried himself called that opinion to the attention of his readers in his New York Review of Books debate with Professor Ronald Dworkin. Fried & Dworkin, supra note 56, at 8 n.4. Indeed, Justice Fried understood what Professor Fried and the Supreme Court's December 12 per curiam majority did not:
[A] voter [who fails to follow directions] should not automatically be disqualified ... like a litigant or one seeking favors from the government, because he or she failed to comply strictly with announced procedures. The voters are the owners of the government, and our rule that we seek to discern the voter's intention and to give it effect reflects the proper relation between government and those to whom it is responsible.
Delahunt, 671 N.E.2d at 1243.
n279. 303 U.S. 444 (1938).
n280. Id. at 450-51.
n281. 486 U.S. 750 (1988).
n282. Id. at 758 (citations omitted). Similarly, a vagrancy statute is void when it is so vague that it ""fails to give a person of ordinary intelligence fair notice that his contemplated conduct is forbidden by the statute,' and ... encourages arbitrary and erratic arrests and convictions," Papachristou v. City of Jacksonville, 405 U.S. 156, 162 (1972) (citations omitted) (quoting United States v. Harriss, 347 U.S. 612, 617 (1954)), or when it "vests virtually complete discretion in the hands of the police to determine whether the suspect has satisfied the statute ... [with the potential to] arbitrarily suppress[] First Amendment liberties," Kolender v. Lawson, 461 U.S. 352, 358 (1983) (citations omitted).
n283. City of Chicago v. Morales, 119 S. Ct. 1849, 1859 (1999). In Morales, Justice Scalia railed against the plurality's efforts to establish loitering as a fundamental liberty protected from discretionary police power by substantive due process. "[The plurality] never does marshal any support for the proposition that loitering is a constitutional right, contenting itself with a (transparently inadequate) explanation of why the historical record of laws banning loitering does not positively contradict that proposition, and the (transparently erroneous) assertion that the city of Chicago appears to have conceded the point." Id. at 1874-75 (Scalia, J., dissenting). Sound familiar? Try substituting "voting rights" for "loitering." See Bush, 121 S. Ct. at 538 (Rehnquist, C.J., concurring) (stating that it is the intent of the legislature, rather than the intent of voters, that matters in presidential elections). Similarly, Justice Thomas's opinion accuses the plurality of indulging the "sweeping conclusion that this ordinance infringes upon a liberty interest protected by the Fourteenth Amendment's Due Process Clause[, a conclusion that] withers when exposed to the relevant history." Morales, 119 S. Ct. at 1881 (Thomas, J., dissenting). Justice Thomas seems particularly suspicious of "judge-made constitutional law having little or no cognizable roots in the language or even the design of the Constitution" and notes that we "should be extremely reluctant to breathe still further substantive content into the Due Process Clause so as to strike down legislation adopted by a State or city to promote its welfare." Id. at 1883. Had the Justice substituted "ability to count ballots to ascertain the intent of its voters" for "welfare," I would have been in complete agreement.
n284. See, e.g., Vill. of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 497-99 (1982).
n285. City of Lakewood, 486 U.S. at 776-78 (White, J., dissenting).
n286. City of Lakewood, 486 U.S. at 761 (rejecting the dissent's analogy between newspaper dispensing devices and soft drink vending machines). The majority held that a city ordinance that vests in a mayor unbridled discretion over which publishers may place newsracks on public property is unconstitutional. Id. at 772.
n287. Bush, 121 S. Ct. at 529.
n288. But see Burdick v. Takushi, 504 U.S. 428, 433-34, 441 (1992) (upholding a state ban on write-in ballots, applying less rigorous scrutiny than would befit a ban on speech, and explaining why voting is not entitled to the protection that free speech receives).
n289. City of Lakewood, 486 U.S. at 760.
n290. Id.
n291. Justice Scalia could not have favored any such extension, given his view that "the entire practice of using the Due Process Clause to add judicially favored rights to the limitations upon democracy set forth in the Bill of Rights (usually under the rubric of so-called "substantive due process') is ... judicial usurpation." City of Chicago v. Morales, 119 S. Ct. 1849, 1873 (1999) (Scalia, J., dissenting).
n292. Burdick, 504 U.S. at 433.
n293. 504 U.S. 428 (1992).
n294. Id. at 436-37.
n295. A voter may not even do that much; the voter may, for example, spoil the ballot to register discontent with the available options.
n296. It is difficult, for example, to gauge what speech the Democrat "Buchanan" voters in Palm Beach made.
n297. Burdick, 504 U.S. at 438 (citation omitted).
n298. Compare id. at 438-39, with Anderson v. Celebrezze, 460 U.S. 780, 792-94, 806 (1983).
n299. Bush, 121 S. Ct. at 530-32.
n300. See City of Lakewood v. Plain Dealer Publ'g Co., 486 U.S. 750, 761-62 (1988) (explaining that such a risk would be insufficient to justify judicial intervention into a discretionary building permit scheme).
n301. Reagan v. Time, Inc., 468 U.S. 641, 648, 653-56 (1984). This case arose when the Secret Service, which for decades had been warning Time, Inc. that its photographic reproductions of currency violated the relevant regulations, seized all plates and materials used in connection with the production of the cover of an issue of Sports Illustrated.
n302. See Transcript of Oral Argument, supra note 100, at 58.
n303. 497 U.S. 261 (1990).
n304. Id. at 289-90 (O'Connor, J., concurring).
n305. 439 U.S. 379 (1979); id. at 383-89.
n306. 414 U.S. 632 (1974); id. at 639-43 (holding that a school board must give pregnant teachers individualized opportunities to show fitness notwithstanding a school rule requiring unpaid maternity leave several months before childbirth).
n307. 405 U.S. 645, 649 (1972) (holding that an unwed father must be given an individualized opportunity to show he is a fit parent).
n308. See, e.g., Saenz v. Roe, 526 U.S. 489, 493, 507 (1999) (overturning a California law limiting welfare payments for new arrivals, until they completed one year of residence in California, to the level offered by their prior state of residence); Vlandis v. Kline, 412 U.S. 441, 452 (1973) (holding that a student who applies to a university from out of state and seeks reduced tuition as a resident must be given an opportunity to prove bona fide resident status); Dunn v. Blumstein, 405 U.S. 330, 359-60 (1972) (overturning state and local durational residency limits for voting).
n309. New York did not introduce mechanical lever machines until the end of the nineteenth century; by 1930 every major metropolitan area had adopted lever machinery. Punchcard machines, using computer technology, did not arrive until the 1960s. See CalTech-MIT Voting Technology Project, Voting: What Is, What Could Be 18 (July 2001), available at http://web.mit.edu/newsoffice/nr/2001/VTP report all.pdf.
n310. See Bush, 121 S. Ct. at 532 ("The recount process, in its features here described, is inconsistent with the minimum procedures necessary to protect the fundamental right of each voter in the special instance of a statewide recount under the authority of a single state judicial officer. Our consideration is limited to the present circumstances, for the problem of equal protection in election processes generally presents many complexities.").
n311. See infra section B.3, pp. 258-63.
n312. Even if it did, this condition would not support a line between the initial process of counting ballots, often manually, and whatever recounting was done in an election so close it called for a recount. The process of manual counting before a result is announced nationwide may occur continuously, there being no constitutional requirement that all the vote counters act simultaneously and in splendid isolation from one another and the rest of the world. See infra note 376. Thus, even in the initial counting process, there is plenty of opportunity for the counter to learn how things look elsewhere.
n313. The following states use an "intent of the voter" standard: Arizona, Connecticut, Indiana, Maine, Maryland, Massachusetts, Michigan, Missouri, Texas, Utah, Vermont, Virginia, Washington, and Wyoming. See Bush, 121 S. Ct. at 540 n.2 (Stevens, J., dissenting). The following states employ a standard in which a vote is counted unless it is "impossible to determine the elector's [or voter's] choice": Alabama, Arizona, California, Colorado, Delaware, Idaho, Illinois, Iowa, Maine, Minnesota, Montana, Nevada, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, South Carolina, South Dakota, Tennessee, and West Virginia. See id. (alteration in original).
n314. See, e.g., Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227 (1995) ("The Fifth and Fourteenth Amendments to the Constitution protect persons, not groups... . [Race is] a group classification long recognized as "in most circumstances irrelevant and therefore prohibited.'" (emphasis omitted) (quoting Hirabayashi v. United States, 320 U.S. 81, 100 (1943))); see also id. at 272-73 (Ginsburg, J., dissenting) (discussing the nation's history of racial oppression).
n315. See, e.g., City of Richmond v. J.A. Croson Co., 488 U.S. 469, 521 (1989) (Scalia, J., concurring in the judgment) ("Our Constitution is color-blind, and neither knows nor tolerates classes among citizens." (quoting Plessy v. Ferguson, 163 U.S. 537, 559 (1896) (Harlan, J., dissenting))). Compare Adarand, 515 U.S. at 229-30 ("Perhaps it is not the standard of strict scrutiny itself, but our use of the concepts of "consistency' and "congruence' in conjunction with it, that leads Justice Stevens to dissent. According to Justice Stevens, our view of consistency "equates remedial preferences with invidious discrimination,' and ignores the difference between "an engine of oppression' and an effort "to foster equality in society,' or, more colorfully, "between a "No Trespassing" sign and a welcome mat.' It does nothing of the kind. The principle of consistency simply means that whenever the government treats any person unequally because of his or her race, that person has suffered an injury that falls squarely within the language and spirit of the Constitution's guarantee of equal protection." (internal citations omitted)), with id. at 245 (Stevens, J., dissenting) ("The consistency that the Court espouses would disregard the difference between a "No Trespassing' sign and a welcome mat.").
n316. Justice Kennedy, although signing and perhaps writing the Bush v. Gore per curiam opinion, wrote for a closely divided Court a few months later that an immigration rule distinguishing on its face between nonmarital children born abroad to mothers who were then United States citizens and nonmarital children born abroad to fathers who were then United States citizens did not violate norms of equal protection. Ironically, he chided his four dissenting colleagues - Justices O'Connor, Souter, Ginsburg, and Breyer - for ignoring the facts of life, including "even our most basic biological differences - such as the fact that a mother must be present at birth but the father need not be - [and thereby risking] making the guarantee of equal protection superficial, and so disserving it." Tuan Anh Nguyen v. INS, 121 S. Ct. 2053, 2066 (2001).
n317. 481 U.S. 279 (1987).
n318. Id. at 286.
n319. Id. at 309.
n320. Id. at 312.
n321. Id. at 315-16.
n322. See Randall L. Kennedy, McCleskey v. Kemp: Race, Capital Punishment, and the Supreme Court, 101 Harv. L. Rev. 1388, 1414 (1988) (quoting McCleskey, 481 U.S. at 339 (Brennan, J., dissenting)).
n323. McCleskey, 481 U.S. at 318.
n324. Dawson v. Delaware, 503 U.S. 159, 173 (1992) (Thomas, J., dissenting). Justice Thomas argued that Dawson's membership in a prison gang provided a reasonable basis for a jury to conclude that he had engaged in forbidden activities in prison. Id. at 172-73.
n325. Dawson, 503 U.S. at 166.
n326. Id. at 169.
n327. Id. at 171-72 (Thomas, J., dissenting).
n328. Richard H. Pildes & Richard G. Niemi, Expressive Harms, "Bizarre Districts," and Voting Rights: Evaluating Election-District Appearances After Shaw v. Reno, 92 Mich. L. Rev. 483, 506-07 (1993) ("On this view, the meaning of a governmental action is just as important as what that action does.").
n329. See, e.g., Richard H. Pildes, Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism, 27 J. Legal Stud. 725, 760 (1998) ("Redistricting might be thought one of the most appropriate sites for judicial concern with expressive harms, for district designs, routinely displayed in newspapers, signal public values in more visible ways than most other state acts. If government endorses an impermissible vision of the role of race in public life, few moments are likely to represent that endorsement more dramatically than the design of democratic institutions.").
n330. Pildes & Niemi, supra note 328, at 509 ("[The concern for expressive harms] explains why the Court is adamant that "reapportionment is one area in which appearances do matter.' If they do, it must be because, even apart from any concrete harm to individual voters, such appearances themselves express a value structure that offends constitutional principles." (footnote omitted) (quoting Shaw v. Reno, 509 U.S. 630, 647 (1993))); see also Peter J. Rubin, Reconnecting Doctrine and Purpose: A Comprehensive Approach to Strict Scrutiny After Adarand and Shaw, 149 U. Pa. L. Rev. 1 (2000).
n331. Indeed, many scholars have noted the importance of appearance in Establishment Clause jurisprudence:
All the hallmarks of an expressive theory are present in Establishment doctrine. First, the doctrine prohibits state infliction of purely expressive harms, even when unaccompanied by differential causal impact between adherents and nonadherents of a religion. Second, far from treating laws with the same effects equally, it treats certain effects as of constitutional concern only when they are caused by a law already found objectionable on expressive grounds. Third, Establishment doctrine clearly distinguishes the expressive meaning of state action from what state actors intended to communicate, and holds the State accountable for the former. Fourth, it recognizes how the State's communications can expressively harm people by changing their relationship to the State.
Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General Restatement, 148 U. Pa. L. Rev. 1503, 1546-47 (2000).
n332. Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819, 841 (1995); see also Mitchell v. Helms, 120 S. Ct. 2530, 2547, 2548 (2000).
n333. County of Allegheny v. ACLU, 492 U.S. 573, 620 (1989) (quoting Laurence H. Tribe, American Constitutional Law 14-15, at 1296 (2d ed. 1988)).
n334. Many respectable journalists have attributed authorship of the per curiam opinion to Justice Kennedy. See, e.g., Kaplan, supra note 13, at 274 ("[The opinion] was obviously the view of only Kennedy or O'Connor or both... . Later it became clear that Kennedy himself had drafted the opinion.").
n335. Shaw v. Reno, 509 U.S. 630, 647 (1993).
n336. 515 U.S. 200 (1995).
n337. Id. at 229 (emphasis in original).
n338. 458 U.S. 718 (1982).
n339. Id. at 725; see also Deborah Hellman, The Expressive Dimension of Equal Protection, 85 Minn. L. Rev. 1, 33 (2000).
n340. Tuan Anh Nguyen v. INS, 121 S. Ct. 2053, 2066 (2001) (O'Connor, J., dissenting).
n341. Id. at 2075-76; see also J.E.B. v. Alabama ex rel T.B., 511 U.S. 127, 146 (1994) (O'Connor, J., concurring) (conceding the probability that female jurors would tend to view the evidence in a paternity suit, or a rape case, differently from male jurors, but observing that the Court's holding requires that we look the other way and act as though this were not so, thereby preserving an appearance of gender-neutrality).
n342. The per curiam proclaimed that its remedy's raison d'<cri e>tre was "to sustain the confidence that all citizens must have in the outcome of elections." Bush, 121 S. Ct. at 532. Justice Kennedy seemed similarly focused on appearances when, dissenting from the Court's 5-4 invalidation of a state ban on so-called "partial-birth abortion," he observed how much the banned procedure, in contrast to procedures the state permitted, physically resembled killing an infant in the process of being born. Stenberg v. Carhart, 120 S. Ct. 2597, 2626 (2000) (Kennedy, J., dissenting).
n343. Anderson & Pildes, supra note 331, at 1547.
n344. Lynch v. Donnelly, 465 U.S. 668, 688 (1984) (O'Connor, J., concurring).
n345. Richmond Newspapers v. Virginia, 448 U.S. 555, 572 (1980).
n346. Shaw v. Reno, 509 U.S. 630, 635-36 (1993).
n347. Bush, 121 S. Ct. at 530.
n348. Shaw, 509 U.S. at 647.
n349. Bush, 121 S. Ct. at 532.
n350. See, e.g., Lund, supra note 262.
n351. Fyodor Dostoevsky, The Brothers Karamazov 257 (Richard Pevear & Larissa Volokhonsky trans., Vintage Books 1991) (1880).
n352. Compare Shaw, 509 U.S. at 635, with Mitchell v. Helms, 120 S. Ct. 2530, 2590-91 (2000) (Souter, J., dissenting) (criticizing the plurality for "treating and external observer's attribution of religious support to the government as the sole impermissible effect of a government aid scheme").
n353. See supra note 65 (defining "undervotes").
n354. For example, in Osceola County, in which twenty-nine percent of residents are Hispanic, ballots were printed only in English. See Mintz & Slevin, supra note 102.
n355. Of the forty-one of Florida's sixty-seven counties that used optical scanners, only twenty-four came equipped with mechanisms to give voters a second chance if the ballot they attempted to cast was rejected by machines as invalid. None of the punchcard balloting machines used in twenty-four counties were equipped with a second-chance capability. These twenty-four counties had a rejection rate of just under 1 in 25 - 3.9 percent. Mintz & Slevin, supra note 102 (noting that "forty-one counties used optical-scan machines, which electronically read ovals colored in by voters. Another 24 counties used punch cards. One used lever machines and another used paper ballots common in the 1800s").
n356. Bush, 121 S. Ct. at 541 (Stevens, J., dissenting).
n357. See Ronald Dworkin, A Badly Flawed Election, N.Y. Rev. Books, Jan. 11, 2001, at 53, 54; Fried & Dworkin, supra note 56, at 9-10.
n358. Bush, 121 S. Ct. at 531.
n359. Id.
n360. See id. at 531-32.
n361. The five-Justice per curiam said its "consideration is limited to the present circumstances," Bush, 121 S. Ct. at 532, a comment it explained in the next sentence by insisting - as though proclaiming the proposition could make it so - that "the question before the Court is not whether local entities, in the exercise of their expertise, may develop different systems for implementing elections. Instead, we are presented with a situation where a state court with the power to assure uniformity has ordered a statewide recount with minimal procedural safeguards." Id. (emphasis added). Justice Souter evidently agreed that this question could be surgically separated from that of local variation in voting systems and practices, for he noted that, while the "Equal Protection Clause does not forbid" the use of different mechanisms that "will have different levels of effectiveness in recording voters' intentions," in the initial vote count, the use of different standards during the recount stage serves "no legitimate state interest" and "appears wholly arbitrary." Id. at 545 (Souter, J., dissenting). Of the seven Justices who recognized that there may well have been a constitutional problem with the recount, only Justice Breyer flatly denied that this election baby could be so neatly split: "In a system that allows counties to use different types of voting systems, voters already arrive at the polls with an unequal chance that their votes will be counted," he observed without contradiction from any other Justice. Id. at 552 (Breyer, J., dissenting). He ended with a seemingly unanswerable conclusion: "I do not see how the fact that this results from counties' selection of different voting machines rather than a court order makes the outcome any more fair." Id. The per curiam opinion was therefore more than slightly misleading when it asserted: "Seven Justices of the Court agree that there are constitutional problems with the recount ordered by the Florida Supreme Court that demand a remedy." Bush, 121 S. Ct. at 533. One of those seven - Justice Breyer - left no doubt that he deemed it incoherent to deal only with the recount's difficulties while treating the underlying count as separable - to make, as it were, the sound of one hand clapping. And he left the clear impression that, as between (a) halting the recount while doing nothing to remedy the problems with the underlying count and (b) letting the recount go on even without addressing its supposed flaws, he would have found the latter choice easier to square with the Constitution. Nor would it even be correct to say that Justice Breyer agreed that the constitutional problems with the recount "demand a remedy." Id. At most, he agreed that, in the circumstances presented, "basic principles of fairness may well have counseled the adoption of a uniform standard to address the problem." Id. at 551 (Breyer, J., dissenting) (emphasis added). If it seems unfair to parse the opinions so closely when they were, after all, written in great haste, consider the fairness of acting in such haste to shut down the democratic process of a sovereign state of the Union.
n362. Fla. Stat. ch. 102.141(4) (2000).
n363. Mintz & Slevin, supra note 102.
n364. Id.
n365. U.S. Comm'n on Civil Rights, Voting Irregularities in Florida During the 2000 Presidential Election ch. 9, at 2 (2001), at http://www.usccr.gov/vote2000/stdraft1/ ch9.htm; id. app. VII, at http://www.usccr.gov/vote2000/stdraft1/ltrpt.htm. See also Katharine Q. Seelye, Divided Civil Rights Panel Approves Election Report, N.Y. Times, June 9, 2001, at A8; Bob Drogin, Rights Panel Finds Fla. Vote "Injustice,' L.A. Times, June 5, 2001, at A1.
n366. Mintz & Slevin, supra note 102.
n367. Id.
n368. Id.
n369. See supra note 361 (discussing Justice Souter's belief that the recount inequalities were impermissible).
n370. Bush, 121 S. Ct. at 550 (Ginsburg, J., dissenting).
n371. Advocates for Vice President Gore raised the issue of the unreliability of the underlying count in the Palm Beach County case before the Supreme Court. See Brief of Respondents Al Gore, Jr. and Florida Democratic Party at 4 n.2, Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471 (2000) (No. 00-836).
n372. See 3 U.S.C. 1 (requiring that all presidential electors be chosen on the Tuesday following the first Monday in November); see also supra note 10 (discussing my proposal for a new special election).
n373. Bush, 121 S. Ct. at 532.
n374. See id. at 545 (Souter, J., dissenting) (finding that "the Equal Protection Clause does not forbid the use of a variety of voting mechanisms within a jurisdiction, even though different mechanisms will have different levels of effectiveness in recording voters' intentions; local variety can be justified by concerns about cost, the potential value of innovation, and so on").
n375. 411 U.S. 1 (1973).
n376. Indeed, the whole business of treating the occasion of the election as defining a "line" capable in principle of dividing the temporal world into a pre-election period, the moment of the election, and a post-election period, is unreal. Federal law requires that the election for presidential electors be held on the same day in every state, 3 U.S.C. 1 (1994), a requirement that the Constitution expressly authorizes Congress to impose, U.S. Const. art. II, 1, cl. 4. But that provision certainly does not say, nor could it, that the election is to be held everywhere at the same moment or even that each state must find a way to conduct the election so that it begins and ends throughout the state at exactly the same instant. This makes far more difference now, in an era of instantaneous and ubiquitous communications coupled with exit polls at enough polling places, scientifically chosen to constitute a highly informative sample, than it made in the horse and buggy days when the Constitution was written and ratified. Among other things, this means that, by late afternoon in the East Coast, which is only midday on the West Coast, significant election results may already be and often are projected and nationally broadcast. This can occur even in a single state like Florida. Readers will recall the controversy surrounding the impact on voters in Florida's westernmost panhandle (when the polls closed in Miami but were still open in the different time zone of the panhandle) from NBC's early and dramatically premature announcement that Gore had taken Florida. The fact that scores of Florida voters (some Bush supporters who had given up and some Gore supporters who had come to feel superfluous) supposedly went home upon hearing that announcement is unfortunate even if the numbers on both sides may have cancelled each other out. Citizen volunteers and local officials, affected by that broadcast announcement and the resulting voter behavior, may, in making spot decisions at the polls, have altered their attitude toward and handling and processing of voters and their ballots, and of absentee ballots as well.
This is no less likely or troubling than partisan manipulation in the recount process. And it is especially troubling from the perspective of those who would like to assume that there exists a meaningful line of some sort between "pre-election" and "post-election" differences in treatment of voters and of ballots, with pre-election differences being impervious to partisan manipulation (inasmuch as no election results are yet in) and with post-election differences being susceptible to partisan manipulation (inasmuch as they occur after the veil of election-result ignorance has been lifted).
n377. The need to address this issue was called to my attention by my colleague, Professor Heather Gerken. Private Communication with Heather Gerken, Assistant Professor of Law, Harvard Law School, in Cambridge, Mass. (June 2001).
n378. Rodriguez, 411 U.S. at 83-85, 92 (1973) (Marshall, J., dissenting).
n379. An examination of the legislative history of the "safe harbor" provision of the Electoral Count Act of 1887, 3 U.S.C. 5 (1994), interestingly reveals that the concern it addressed was that state legislatures, not courts, would rig new rules after the election had gone by in order to get their favorite into office. Representative Cooper, representing the majority view, argued that it would:
be unwise to permit a legislature to assemble and permit the dominant party, in view of the very existing affairs, in view of the peculiar phases of the contest which is being made, to enact laws governing and deciding that contest - possibly with a view of having it decided in accordance with their wishes rather than with the expressed wish of the people or with justice and the right.
18 Cong. Rec. 47 (1886) (statement of Rep. Cooper). Representative Herbert added:
I understand the House Committee having this bill in charge are practically unanimous, nearly all being in favor of decisions by State tribunals; but the minority say they want to reserve to the States the right to pass a law even after an election. Mr. Speaker, to me this proposition seems mischievous in the extreme.
Id. at 75 (1886) (statement of Rep. Herbert). These statements show that Congress was primarily concerned with post-election measures enacted by state legislatures. The thrust of the congressional debate thus undercuts Chief Justice Rehnquist's focus in his concurrence on post-election modifications by the state judiciary. See Bush, 121 S. Ct. at 539 (Rehnquist, C.J., concurring). There was some concern about post hoc election tribunals constituted to sway the election. See 18 Cong. Rec. 75 (1886) (statement of Rep. Herbert) ("It would simply give the power to any State, after an election was held and a dispute had arisen, to trump up in that State an electoral commission to decide that question according to rules to be made for the occasion, which would enable it to reach precisely the decision desired by the majority in the State Legislature."). But this concern was clearly directed at tribunals created by the state legislature, not at the existing state judiciary. The debates do not indicate any similar concern that state court judges would alter the laws to reach a result more palatable to themselves. This is not to say that 3 U.S.C. 5 does not constrain state judiciaries; rather, it simply demonstrates that the primary concern of those who passed the law was post-election actions taken by the legislature.
n380. See Bush, 121 S. Ct. at 533.
n381. See id. at 533-35 (Rehnquist, C.J., concurring).
n382. Id. at 538 (emphasis added).
n383. Id. at 538-39.
n384. 3 U.S.C. 5 (1994). The Florida election code establishes a unitary system of law to govern all elections and draws absolutely no distinction between presidential elections, with their special federal time limits, and state elections, which contain no such limits. See, e.g., Fla. Stat. ch. 102.166 (2000) (provisions for conducting an election protest for all elections); id. ch. 102.168 (provisions for conducting an election contest for all elections). In fact, references to the federal presidential election merely establish that this code governs the selection of electors in such an election. See Fla. Stat. ch. 97.021 (2000); id. ch. 103.011. This unitary nature of the Florida code makes it impossible for the state legislature ever to have intended for presidential elections to be governed by the 3 U.S.C. 5 or 7 deadlines, which clearly apply only to federal elections. Moreover, the Florida Legislature conferred upon the Florida Supreme Court the power to interpret all of the election code. The legislature's enactment of the election code and its amendments conferred upon the Florida Supreme Court the power to fill in all statutory gaps and resolve any inconsistencies that arose in a state election contest. The role played by the Florida Supreme Court in the presidential contest in interpreting code chapters was identical to its role in state elections. See Reply Brief of Respondents Al Gore Jr. and Florida Democratic Party at 15-16, Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471 (2000) (No. 00-836). Justice Souter also noted at the first oral argument that it was natural for the Florida Legislature to assume that the Florida Supreme Court would play the same role vis-a-vis the meaning of the code provisions as applied to presidential election protests and contests as the court was to play in discerning the meaning of identical code provisions as applied to elections for state offices. See Transcript of Oral Argument at 56-57, Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471 (2000) (No. 00-836), available at http://www.c-span.org/campaign2000/Florida/00-836.pdf ("You have here a statute as I understand it that regulates both Federal and state recounts, that much is, I think is clear ... . The only way to avoid an unconstitutional meaning of the statute so far as Florida law was concerned was to get into this constitutional concern about preserving the franchise, and that because the legislature intended one standard to cover both Federal and state recounts, it therefore is valid to consider the state constitution in order to derive a general meaning that will apply to a Federal, as well as a state election. Can you look at it that way?"). State elections, of course, have no analogous deadlines. Clearly, therefore, it is absurd to imagine that, in crafting a code to cover state and federal, nonpresidential and presidential, elections alike, the Florida Legislature would have implicitly imposed a drop-dead deadline keyed to a date of December 12, December 18, or January 6. If the legislature contemplated grafting any such deadlines, applicable to presidential elections only, onto the general election code, it would necessarily have enacted explicit language to that effect.
n385. See supra note 384; cf. Oregon v. Mitchell, 400 U.S. 112 (1970) (holding that Congress could lower the voting age to eighteen in federal but not state elections). The disparate treatment of federal and state elections in Mitchell engendered such chaos and controversy that it virtually ensured the passage of the Twenty-Sixth Amendment which, one year later, lowered the voting age to eighteen in all elections.
n386. Richard A. Serrano & Ricardo Alonso-Zaldivar, Democrats: Gore to Bow Out if Justices Deal a Defeat, L.A. Times, Dec. 11, 2000, at A25 ("However the U.S. Supreme Court rules, Bush still enjoys support from the Republican-controlled Florida Legislature, where House and Senate committees will convene today to continue a special session ... . The special session is designed to give Florida's 25 electoral college votes to Bush, even if Gore prevails in court.").
n387. See supra note 33.
n388. There is no constitutionally prescribed method by which one Congress may require a future Congress to interpret or discharge a constitutional responsibility in any particular way. See 1 Tribe, supra note 123, 2-3, at 125-26 n.1.
n389. Bush, 121 S. Ct. at 550 (Ginsburg, J., dissenting).
n390. David Kaplan reports that Justice Souter, in a talk to a group of prep-school students a month after the decision, thought aloud that if he'd had "one more day - one more day," he could have brokered a deal with Justice Kennedy to remand the case to Florida and let the political branches sort out the mess. Kaplan, supra note 13, at 284-85.
n391. There is a Catch-22 worthy of note here. As the Court could not deny, the Florida Supreme Court's December 8 recount order, directing all the canvassing boards to apply a general "voter's intent" standard and not any statewide set of uniform sub-standards or rules, carried out the electoral ground rules that the people of Florida, through their legislature, had put in place before election day. Whatever might be said for enforcing some sort of bright-line rule for recounting ballots - right up to the most rigorous ("no punch card ballots count as votes unless one and only one square has been cleanly and fully punched through in accord with the voting instructions") - the text, the context, and the history of Florida's election code make plain that the Florida Legislature had opted for an "intent of the voter" standard instead. Recall the insistent rallying cry against the Florida court: Article II, Section 1, Clause 2 makes the rules laid down by each state legislature before the election the binding guide for choosing presidential electors. The "fix" proposed by the Court, but then withheld because the clock had supposedly run out, and urged by Justices Souter and Breyer for real - a fix in which uniform statewide criteria for counting votes would, time permitting, govern the recount process - would have required adopting new rules in mid-election, one of the cardinal sins that Bush accused the Florida Supreme Court of having committed. So there was an irreducible, if little-noted, tension between the universally accepted importance of judging an election's results according to standards and methods adopted in advance, and the equal protection theory that attracted seven Justices.
n392. Bush, 121 S. Ct. at 541 (Stevens, J., dissenting).
n393. Bush, 121 S. Ct. at 532.
n394. See, e.g., Lani Guinier, A New Voting Rights Movement, N.Y. Times, Dec. 18, 2000, at A27.
n395. See B.J. Palermo, Bush-Gore Lives On, Nat'l L.J. Sept. 17, 2001, at A1 (discussing litigation pending in California, Georgia, Illinois, and Missouri all alleging that punchcard voting systems violate "the equal protection clause of the 14th Amendment by reducing users' chances of having their votes counted accurately").
n396. Lambert v. California, 355 U.S. 225, 232 (1957) (Frankfurter, J., dissenting).
n397. Reynolds v. Sims, 377 U.S. 533 (1964) (invalidating an Alabama apportionment scheme that gave rural voters more influence than urban voters in electing state representatives).
n398. Harper v. Va. State Bd. of Elections, 383 U.S. 663 (1966) (invalidating a Virginia poll tax on equal protection grounds).
n399. 401 U.S. 371 (1971) (holding that Connecticut's law violated due process by requiring an individual to pay a fee as a condition of access to the state's divorce courts); cf. M.L.B. v. S.L.J., 519 U.S. 102 (1996) (holding that equal protection and due process forbid a state from conditioning a parent's right to appeal the termination of parental rights on the parent's ability to pay record transcript costs from the initial termination proceeding); Little v. Streater, 452 U.S. 1 (1981) (holding that due process and equal protection require the state to pay for a blood test at the request of the defendant in a paternity suit).
n400. See supra note 376.
n401. See supra section IV.B.3, pp. 258-63.
n402. See supra pp. 258-60 and note 361.
n403. Bush, 121 S. Ct. at 532; cf. Bendix Autolite Corp. v. Midwesco Enters., 486 U.S. 888, 897 (1988) (Scalia, J., concurring) (comparing meaningless balancing between things that are not commensurable to "judging whether a particular line is longer than a particular rock is heavy").
n404. Bush, 121 S. Ct. at 532.
n405. Cf. Planned Parenthood v. Casey, 505 U.S. 833, 866 (1992) ("The Court's legitimacy depends on making legally principled decisions under circumstances in which their principled character is sufficiently plausible to be accepted by the Nation.").
n406. Cf. Frederick Schauer, Precedent, 39 Stan. L. Rev. 571, 589 (1987) ("The conscientious decisionmaker must recognize that future conscientious decisionmakers will treat her decision as precedent, a realization that will constrain the range of possible decisions about the case at hand.").
n407. Cf. County of Washington v. Gunther, 452 U.S. 161, 183 (1981) (Rehnquist, J., dissenting) (quoting Smith v. Allwright, 321 U.S. 649, 669 (1944) (Roberts, J., dissenting)) (lambasting the majority's narrowly written opinion as an unreasoned decision that "will be treated like a restricted railroad ticket").
n408. Proponents of the Article III argument that all decisions must have precedential effect have focused recently on the circuit rules governing unpublished decisions. Under these rules, unpublished decisions are generally accorded precedential value only under very limited circumstances. Thus an opinion's precedential value will depend on its publication status. The constitutionality of these circuit rules has recently been called into question. In Anastasoff v. United States, 223 F.3d 898 (8th Cir.), vacated on other grounds on rehearing en banc, 235 F.3d 1054 (8th Cir. 2000), Judge Richard S. Arnold held the Eighth Circuit's rule prohibiting citation to unpublished opinions to be a violation of Article III. Id. at 899. This decision has ignited a heated legal debate. For an early and thoughtful discussion of Anastasoff and the issues surrounding it, see Recent Case, 114 Harv. L. Rev. 940 (2001).
n409. See, e.g., Schauer, supra note 406, at 595-601 (arguing that reliance on precedent promotes fairness, predictability, better decisionmaking, and stability).
n410. Certain areas of the law, such as statutes of limitations, have lines of precedent with a very tight fit, or a strong correlation. For these areas of the law, it is relatively easy to tell in any given case how a court will decide. In other areas of law, such as the Court's Establishment Clause and state action jurisprudence, the data points are scattered and the fit of the line connecting them is very loose. For these areas of the law, determining how a given court will resolve a case is decidedly harder.
n411. The Supreme Court itself has acknowledged this fact by granting its own dispositions varying degrees of precedential effect. See, e.g., Caban v. Mohammed, 441 U.S. 380, 390 n.9 (1979) (stating that summary dispositions are not entitled to the "same deference" as full dispositions); Colo. Springs Amusements, Ltd. v. Rizzo, 428 U.S. 913, 916 (1976) (Brennan, J., dissenting from denial of certiorari) ("We accord summary dispositions less precedential value than dispositions by opinion after full briefing and oral argument ... .").
n412. Indeed, Justices Breyer and Stevens, two of the Bush v. Gore dissenters, cited admiringly by critics of the Court's narrow holding, both either brought about or urged similarly limited holdings later in the Court's same Term. Justice Breyer, joined by Justice O'Connor, penned the defining opinion for a divided Court in Bartnicki v. Vopper, 121 S. Ct. 1753, 1766 (2001) (Breyer, J., concurring), in which he agreed "with the Court's holding that the statutes as applied here violate the Constitution," but declined to "extend that holding beyond these present circumstances." Id. at 1769. For Justice Stevens's similar advocacy of limited holdings, see Kyllo v. United States, 121 S. Ct. 2038, 2047 (2001) (Stevens, J., dissenting). Joined by the Chief Justice, Justice Kennedy, and Justice O'Connor, Justice Stevens argued that the majority "failed to heed the tried and true counsel of judicial restraint," in that "instead of concentrating on the rather mundane issue ... actually presented by the case before it, the Court has endeavored to craft an all-encompassing rule for the future." Id. at 2052.
n413. See, e.g., Thornburgh v. Am. Coll. of Obstetricians & Gynecologists, 476 U.S. 747, 819-21 (1986) (O'Connor, J., dissenting). Thus, critics of "judicial legislation" and "activism" sometimes praise the Court as suitably modest when it limits its holdings as narrowly as possible. See, e.g., Cass R. Sunstein, One Case at a Time: Judicial Minimalism on the Supreme Court (1999). At the same time, critics of an activist Court feel free to excoriate decisions that do just that. Chief Justice Burger criticized the decision in Plyler v. Doe, 457 U.S. 202, 243 (1982), for "resting on such a unique confluence of theories and rationales that it will likely stand for little beyond the results in these particular cases." Id. at 243 (Burger, C.J., dissenting). Romer v. Evans, 517 U.S. 620 (1996), is sometimes attacked for the narrow compass of its novel per se invalidation of a statewide constitutional amendment rendering an entire group of citizens "strangers to [the] laws." Id. at 635.
n414. Cf. Willard Van Orman Quine, From A Logical Point of View: Nine Logico-Philosophical Essays (2d ed., rev. 1980); W.V. Quine, The Ways of Paradox and Other Essays (1976); Willard Van Orman Quine, Word and Object (1960).
n415. E.g., Shaffer, supra note 193, act 1, sc. 7, at 21; id. act 2, sc. 2, at 53; id. act 2, sc. 7, at 64.
n416. See supra section IV.A.1, pp. 222-26.
n417. See supra section IV.A.2, pp. 226-33.
n418. See supra section IV.A.3, pp. 233-47.
n419. This view was perhaps best expressed by one conservative pundit, who stated:
The final vote on Election Day, after all, was made up of millions of small contingencies and glitches that weighed back and forth on each side of the ledger - bad ballots, lost ballots, lazy voters, dopey voters, mistakes, and misfortune. There is no way to remove error from human endeavor. Life is chaotic, which is why we need rules to channel it, to give order to happenstance, and keep things from reeling out of control.
This is why we have and need rules for elections, which are standard and neutral and fixed ahead of time. Elections happen on one day, not 12, and not 90. Polls are open during specified hours, and if you are late, it's unfortunate. You get to vote once, not over and over. All votes are counted by one standard, not some once and some four times, with selected votes being subject to interpretation. All voters are equal, and all ballots are equal, or else there is chaos. When rules are chaotic - when deadlines are stretched and at last become meaningless, when votes are read differently, when standards have vanished - then chaos is everywhere. Chaos is everywhere in Florida now, and Republicans have an aversion to this that is not simply partisan.
Emery, supra note 188, at 24-25.
n420. Kaplan, supra note 13, at 293. Kaplan quotes Justice Kennedy as saying "Sometimes you have to be responsible and step up to the plate ... you have to take responsibility." Id.
n421. See Ely, supra note 184.
n422. See William J. Brennan, The 1986 Oliver Wendell Holmes, Jr. Lecture: Constitutional Adjudication and the Death Penalty: A View from the Court, 100 Harv. L. Rev. 313 (1986).
n423. Bush, 121 S. Ct. at 533.
n424. Id. (emphasis added).
n425. Id.
n426. Id.
n427. Gore v. Harris, 772 So. 2d 1243, 1270 (Fla. 2000) (Wells, C.J., dissenting) (quoting John Allen Paulos, Professor of Mathematics, Temple University).
n428. Id. at 1263.
n429. See Bush, 121 S. Ct. at 529-30.
n430. 3 U.S.C. 2 (1994) (emphasis added).
n431. See U.S. Const. amend. XII ("The President of the Senate shall, in the presence of the Senate and House of Representatives, open all the certificates and the votes shall then be counted ... .").
n432. See supra note 388.
n433. I have long argued that the Constitution vests in Congress the primary responsibility for resolving presidential election disputes. On the eve of the 1980 presidential election among Jimmy Carter, Ronald Reagan, and John Anderson, Thomas Rollins and I discussed a set of possible outcomes similar to those advanced in Judge Posner's book (a bitter congressional battle, attempts at nefarious deal-making, and even the installation of an interim President). I argued then, as I do now, in favor of allowing democratic processes to run their course as prescribed by the Constitution, recognizing that, "in truth, what we have learned to live with in the presidential selection system is a complicated process, rich in possibilities for high statesmanship and low politics alike." Laurence H. Tribe & Thomas M. Rollins, Deadlock: What Happens if Nobody Wins, Atlantic Monthly, Oct. 1980, at 49, 61. Evidently - and before Bush v. Gore found its way into the Supreme Court's doors - other observers shared my conclusion. For instance, in the first round of the litigation, Professors Charles Fried and Einer Elhauge made a forceful argument that the case presented a nonjusticiable political question. See Brief of the Florida Senate and House of Representatives as Amici Curiae in Support of Neither Party at 7, Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471 (2000) (No. 00-836); see also Chemerinsky, supra note 232, at 1105-09.
n434. See U.S. Const. amend. XII (declaring that the President of the Senate shall count the electoral votes "in the presence of the Senate and House of Representatives" and that, if no candidate for President receives a majority of the votes of the electors, "the House of Representatives shall choose immediately, by ballot, the President"); Guido Calabresi, In Partial (but Not Partisan) Praise of Principle, in The Question of Legitimacy, supra note 4, at 79, 80 (arguing that one approach to Bush v. Gore would have been to "let the House and Senate do their job"); id. at 87 ("On what basis, this criticism would add, can an unelected priesthood assert that it can define what is and what is not fair in so profoundly populist and political an event as an election?"); Elizabeth Garrett, Leaving the Decision to Congress, in The Vote, supra note 3, at 38, 48-54.
n435. See Baker v. Carr, 369 U.S. 186, 217 (1962).
n436. U.S. Const. amend. XII.
n437. 424 U.S. 1 (1976).
n438. See INS v. Chadha, 462 U.S. 919, 959 (1983).
n439. Also relevant to the selection of President is the Twelfth Amendment's provision that, in the event no candidate receives a majority of the electoral votes after counting, "the House of Representatives shall choose immediately, by ballot, the President." U.S. Const. amend. XII.
n440. H.R. Rep. No. 49-1638, at 2 (1886) (as submitted by Rep. Caldwell for the Select Committee on the Election of President and Vice-President).
n441. Baker v. Carr, 369 U.S. 186, 215 (1962).
n442. The message was put squarely before the Court in the amicus briefs for the state legislature. See Brief of the Florida House of Representatives and Florida Senate, supra note 278, at 2; Brief of the Florida Senate and House of Representatives, supra note 433, at 7 (amicus brief of Charles Fried and Einer Elhauge).
n443. Bush, 121 S. Ct. at 555 (Breyer, J., dissenting). See also Bush, 121 S. Ct. at 545 (Souter, J., dissenting) (arguing that "if not disposed of at the state level [the election contest] could have been considered by the Congress in any electoral vote dispute").
n444. See Alexander M. Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics 111-98 (1962).
n445. Charles Lane, 2 Justices Defend Court's Intervention in Fla. Dispute, Wash. Post, Mar. 30, 2001, at A13; see also Justices Defend Election Involvement, N.Y. Times, Mar. 30, 2001, at A1.
n446. Bush, 121 S. Ct. at 556 (Breyer, J., dissenting) (quoting Speech of James Madison at the Federal Convention (July 25, 1787), in 5 Debates on the Adoption of the Federal Constitution 363 (Jonathan Elliot ed., 1987)) (internal quotation marks omitted).
n447. Id.
n448. Id.
n449. See infra Part VI, pp. 292-99.
n450. U.S. Const. art. I, 3, cl. 6.
n451. Nixon v. United States, 506 U.S. 224, 229-38 (1993).
n452. Id.
n453. See supra section IV.A.4, pp. 247-54.
n454. See infra p. 298.
n455. For an example, see 1 Tribe, supra note 123, 3-13, at 376, which compares Gilligan v. Morgan, 413 U.S. 1 (1973), in which the adequacy of the training of the Ohio National Guard after the Kent State massacre was deemed a political question because the relief sought - an injunction against the Ohio National Guard - was a federal judicial takeover of the training process, which the Constitution had committed to the military and to Congress, id. at 11-12, with Scheuer v. Rhodes, 416 U.S. 232 (1974), in which the same "question" - the adequacy of the National Guard's training - was treated as justiciable in a ruling denying the state governor immunity from liability for the shootings, id. at 250.
n456. Not uncommonly, the same dispute will be vulnerable to a defense of nonjusticiability both on the ground that it is not ripe for decision by an Article III court and on the ground that, in any event, it is nonjusticiably political. See, e.g., Goldwater v. Carter, 444 U.S. 996, 1002-03 (1979) (Rehnquist, J., concurring in the judgment) (finding that Senator Barry Goldwater's challenge to President Carter's unilateral termination of a treaty with Taiwan was a nonjusticiable political question); id. at 997-98 (Powell, J., concurring) (arguing that the decision was not ripe for review because neither the House nor the Senate had created a "constitutional impasse" by challenging the President's decision).
n457. 146 U.S. 1, 35 (1892) (upholding a state legislature's authority to partition a state into election districts, with each district selecting one or more electors, rather than adopting the more common system of selecting the entire slate of electors by a single, at-large vote of the state's citizens).
n458. Id. at 23-24.
n459. Id. at 23.
n460. 418 U.S. 683 (1974).
n461. Id. at 692-93.
n462. Id. at 697.
n463. Posner, supra note 66, at 1.
n464. After describing a "worst-case scenario that is by no means fantastic, or even highly improbable," id. at 7, Posner intones, "Had that scenario been enacted, the forty-third President would have taken office after long delay, with no transition, with greatly impaired authority, perhaps amid unprecedented partisan bickering and bitterness, leaving a trail of poisonous suspicion of covert deals and corrupt maneuvers, and after an interregnum unsettling to the global and the U.S. domestic economy and possibly threatening to world peace," id. at 18. See also id. at 16 ("Had the Court decided not to stop the Florida recount, there is very little that I can see that would have headed off the worst-case scenario that I sketched earlier. There was nowhere else to shop for a resolution of the deadlock. The scenario was not inevitable; but given the weirdness of recent U.S. political competition, it could not be thought phantasmal.").
n465. Id. at 18.
n466. Id. at 17.
n467. Id. at 23.
n468. Id. at 24.
n469. Id. at 26-28. Perhaps it goes without saying that, although I agree with Judge Posner that under Article II, "the Supreme Court is authorized to intervene if, in the guise of interpretation, the state courts in effect rewrite the state election law," id. at 26, I disagree (for all the reasons described at Part II, pp. 184-94, above) with his conclusion that invalidation of the recount the Florida Supreme Court ordered was "a plausible application of Article II," id. at 27.
n470. See id. at 27-28. According to Judge Posner, reliance on Article II rather than the Equal Protection Clause would have enabled the Justices to elude much of the criticism subsequently leveled against the per curiam opinion. For example, "there would have been no need for the opinion to say that the decision had no precedential effect, since the ground of the decision would have had no implications for election administration generally; [and] the Article II ground, being esoteric, would not have provided a handle for criticisms that the general public could understand." Id. at 27.
n471. 323 U.S. 214 (1944).
n472. Posner, supra note 56, at 20.
n473. One virtue of the Korematsu decision that Judge Posner's analysis shares but the per curiam opinion in Bush v. Gore conspicuously lacks is a frank acknowledgment and candid discussion of the alleged "practical exigencies" shaping the Court's analysis. Yet Judge Posner does not criticize the Bush Court for duplicity, nor does he ever explain how the pragmatism he advocates could justifiably allow the Court to weigh the risk of global instability during an interim presidency while declining to weigh the global and domestic risks and benefits of a Gore presidency versus a Bush presidency, a balance that Posner, like just about everyone else, would think it inappropriate for the Court to strike, whether openly or covertly.
n474. Bush, 121 S. Ct. at 512 (Scalia, J., concurring).
n475. Cf. Bork, supra note 64, at 6 ("We are entitled to speculate that, with good reason, the majority so distrusted the recount process in Florida and the state's courts that it seemed better to end the matter on the twelfth rather than put the country and the court through six more days of legal chicanery and useless turmoil.").
n476. The image of Hercules is invoked with apologies to Ronald Dworkin, who never confused the ideal of the omniscient Hercules with the reality of judges as mere mortals. See Ronald Dworkin, Hard Cases, 88 Harv. L. Rev. 1057, 1083-1109 (1975).
n477. One friend, Dean John Roberts of DePaul College of Law, encapsulated that kind of response for me when he wrote:
I still get angry at the critics who made fun of the attempts of local officials to determine whether a ballot had been punched through by holding it up to the light. I jumped up at the time and yelled at the television set, this isn't funny! That is someone's preference for President and we should try to find out what it says!
E-mail from John Roberts, Dean Emeritus and Professor of Law, DePaul College of Law, to Laurence H. Tribe, Professor of Law, Harvard Law School (Sept. 21, 2001, 11:39 AM) (on file with the Harvard Law School Library).
n478. 520 U.S. 351 (1997).
n479. See Michael W. McConnell, The Supreme Court, 1996 Term - Comment: Institutions and Interpretation: A Critique of City of Boerne v. Flores, 111 Harv. L. Rev. 153, 171-73, 185-88 (1997).
n480. See 1 Tribe, supra note 123, 5-16, at 949-51 n.121, 998-99, 5-19, at 998-99 n.74.
n481. See Robert C. Post & Reva B. Siegel, Equal Protection By Law: Federal Antidiscrimination Legislation after Morrison and Kimel, 110 Yale L.J. 441, 521 (2000).
n482. See City of Boerne v. Flores, 521 U.S. 507, 536 (1997).
n483. See Fla. Prepaid Postsecondary Educ. Expense Bd. v. Coll. Sav. Bank, 119 S. Ct. 2199, 2210-11 (1999).
n484. See Coll. Sav. Bank v. Fla. Prepaid Postsecondary Educ. Expense Bd., 119 S. Ct. 2219, 2233 (1999).
n485. See Kimel v. Fla. Bd. of Regents, 120 S. Ct. 631, 650 (2000).
n486. See Bd. of Trs. of Univ. of Ala. v. Garrett, 121 S. Ct. 955, 960 (2001).
n487. 120 S. Ct. 2326 (2000).
n488. 384 U.S. 436 (1966).
n489. See Yale Kamisar, Foreword: From Miranda to 3501 to Dickerson to ..., 99 Mich. L. Rev. 879, 885-89 (2001).
n490. Dickerson, 120 S. Ct. at 2336.
n491. Id. at 2334.
n492. Id. at 2335-36.
n493. 505 U.S. 833 (1992).
n494. 410 U.S. 113 (1973).
n495. See Casey, 505 U.S. at 866-69.
n496. Id. at 868.
n497. Mary Ann Glendon, A Nation Under Lawyers: How the Crisis in the Legal Profession Is Transforming American Society 112 (1994) (quoting Terry Carter, Crossing the Rubicon, Cal. Law., Oct. 1992, at 39, 40) (internal quotation marks omitted).
n498. The one exception was Dickerson, in which the Chief Justice wrote for the Court and Justices Scalia and Thomas dissented.
n499. A similar conclusion is reached in Larry D. Kramer, The Supreme Court, 2000 Term - Foreword: We the Court, 115 Harv. L. Rev. 4 (2001). For an intriguing defense of Bush v. Gore predicated on similar assumptions, see John C. Yoo, In Defense of the Court's Legitimacy, 68 U. Chi. L. Rev. 775 (2001). Professor Yoo turns the left's growing dismay with the aggressive activism of the currently conservative federal judiciary on its head. The fact that the Court has consistently been willing to step on toes, he says, actually grants the Court's decision in Bush v. Gore greater legitimacy. The Court has believed its own press for so long (or at least, according to Yoo, since Planned Parenthood v. Casey), thinking itself the sole appropriate (if avowedly reluctant) arbiter of national controversy, that the Court's intrusion in Bush v. Gore was to be expected as one in a long line of interventions into national debate. And because "consistency of action constitutes a core feature of theories about the Court's legitimacy," the Court's foray into "the Bush" should boost, not deplete, the Court's cachet. Id. at 790. As I explain in this section and later at pp. 302-04, below, to use Bush v. Gore in this way would be a gross perversion of our nation's most celebrated constitutional traditions.
n500. See Scalia, supra note 79, at 1175.
n501. See Brown v. Bd. of Educ., 347 U.S. 483, 495 (1954).
n502. See Bickel, supra note 444, at 264-65; Post & Siegel, supra note 481, at 521.
n503. United States v. Nixon, 418 U.S. 683, 713-14 (1974).
n504. 5 U.S. (1 Cranch) 137 (1803).
n505. See 1 Tribe, supra note 123, 3-2, at 211-12.
n506. The Court itself served as an example for Congress, even if only momentarily, through the obvious intelligence of its oral proceedings of December 1 and the statecraft reflected in its unanimous remand of December 4. See Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 471, 475 (2000). There, the Court implicitly rebuffed the theory that the Florida Supreme Court had fatally interfered with what Bush claimed was the state legislature's exclusive Article II, Section 1, Clause 2 authority, and had violated 3 U.S.C. 5. The Court rejected that theory by giving the Florida court a chance to clarify its stance and potentially to reinstate its November 21 decision, see id., which the Florida court proceeded to do, see Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1273 (Fla. 2000).
n507. See, e.g., Bugliosi, supra note 2; Dershowitz, supra note 2.
n508. Although some commentators seem to believe that the Court's December 12 decision was a solid 7-2 majority as to its equal protection holding, the actual decision to halt Florida's electoral processes was certainly a much weaker 5-4. Justice Souter's dissent did contain dictum endorsing the substantive equal protection theory of the per curiam opinion, but it is misleading to say that either Justice Souter or Justice Breyer, whose dissent contained a tepid acceptance of the majority's equal protection theory, agreed with the per curiam opinion's resolution of the case. See Bush, 121 S. Ct. at 545 (Souter, J., dissenting) (arguing that, although the equal protection issue alone presented a "meritorious argument for relief ... . it is an issue that might well have been dealt with adequately by the Florida courts if the state proceedings had not been interrupted, and if not disposed of at the state level it could have been considered by the Congress in any electoral vote dispute"); id. at 551 (Breyer, J., dissenting) (describing the majority's equal protection theory and equivocally stating, "I agree that, in these very special circumstances, basic principles of fairness may well have counseled the adoption of a uniform standard to address the problem").
The core disagreement between Justices Souter and Breyer and the remaining five Justices focused on the appropriate remedy for any equal protection injury. The five-Justice majority's remedy fatally undermined the equal protection theory, see supra pp. 264-65, and both Justice Souter and Justice Breyer dissented because the remedy the majority adopted was fundamentally inconsistent with the merits of the equal protection principles they endorsed:
By halting the manual recount, and thus ensuring that the uncounted legal votes will not be counted under any standard, this Court crafts a remedy out of proportion to the asserted harm. And that remedy harms the very fairness interests the Court is attempting to protect. The manual recount would itself redress a problem of unequal treatment of ballots. As Justice Stevens points out, the ballots of voters in counties that use punch-card systems are more likely to be disqualified than those in counties using optical-scanning systems. According to recent news reports, variations in the undervote rate are even more pronounced. Thus, in a system that allows counties to use different types of voting systems, voters already arrive at the polls with an unequal chance that their votes will be counted. I do not see how the fact that this results from counties' selection of different voting machines rather than a court order makes the outcome any more fair.
Id. at 552 (citations omitted). Justice Breyer and Justice Souter concluded that the only proper remedy, consistent with the equal protection views of the five-Justice majority, was to remand to the state courts for the development and application of a uniform counting standard for the several types of ballots used in Florida. See id. at 545-46 (Souter, J., dissenting) ("Unlike the majority, I see no warrant for this Court to assume that Florida could not possibly comply with this requirement before the date set for the meeting of electors, December 18... . There is no justification for denying the State the opportunity to try to count all disputed ballots now."); id. at 551 (Breyer, J., dissenting) ("There is no justification for the majority's remedy, which is simply to reverse the lower court and halt the recount entirely. An appropriate remedy would be, instead, to remand this case with instructions that, even at this late date, would permit the Florida Supreme Court to require recounting all undercounted votes in Florida.").
n509. To be sure, one of those two moderates (Justice Breyer) accepted the per curiam's equal protection theory only tentatively, and neither of them agreed with the per curiam's remedial conclusion that a proper recount would have to be completed by midnight on the day of the Court's decision - a deadline then two hours away. Even given those Justices' disagreements with the majority, both still agreed, at the very least, that the Florida Supreme Court's initial recount order probably violated the Equal Protection Clause. To say, then, that the equal protection argument itself was transparently bogus - so bogus that it could have been endorsed only by a jurist whose wish to see Bush prevail overcame all scruples about signing on to a ridiculous argument - ignores the fact that two Justices with little to gain from a Bush victory affirmed the basic theory behind the per curiam opinion. Evidently there was something more at play here than a bald partisan effort to anoint a Republican President.
n510. Indeed, critics claiming partisan bias hardly attempt to account for other explanations for the Court's badly flawed ruling: the rush of time; susceptibility to superficially appealing but ultimately untenable arguments; a deep distrust of Congress; an excessive faith in the Court's own wisdom and an exaggerated view of its mandate; an intractable distaste for apparent disorder; fear of political turmoil; or, just conceivably, something showing that I am - heaven forbid - just plain wrong on the merits. See, e.g., Jeffrey Rosen, Disgrace, New Republic, Dec. 25, 2000, at 18, 18 (criticizing the majority for ending citizens' "faith in the rule of law as something larger than their self-interested political preferences"); Editorial, Unsafe Harbor, New Republic, Dec. 25, 2000, at 9, 9 ("This ruling was designed to bring about a political outcome, and it is an insult to the intelligence of the American people to suggest otherwise.").
n511. See, e.g., Bugliosi, supra note 2; Dershowitz, supra note 2.
n512. See supra pp. 276-77.
n513. A similar argument - with corresponding difficulties of proof - is that partisan motives drove the Court to "bait and switch" the Florida Supreme Court. Some commentators think that the Court deliberately deferred consideration of the equal protection claims initially presented, both in the petition for certiorari to the Florida Supreme Court on November 23 and in the petition for certiorari before judgment in the Eleventh Circuit on the same date, so that, when the claim was finally considered, it would be too late to implement a remedy. See, e.g., Michael C. Dorf, Commentary, Supreme Court Pulled a Bait and Switch, L.A. Times, Dec. 14, 2000, at B11 (arguing that the U.S. Supreme Court trapped the Florida Supreme Court by refusing to decide the equal protection issues in the initial phase of the election, and then remanding on the Article II issue in such a way that the Florida court would be timid about setting uniform statewide standards for the recount, as Gore had urged it to do). These commentators also suggest that, by brandishing the sword of Article II, the Court's unanimous per curiam opinion of December 4 laid a "trap" for the Florida Supreme Court. In effect, the U.S. Supreme Court is said to have implied that the Florida Supreme Court had better not lay down uniform standards lest its entire recount effort be doomed for violating Article II by deviating from the Florida legislature's deliberately vague and general "intent of the voter" standard. See id. Then, when on December 8 the Florida court took the bait and left the state legislature's standard in place, it was slapped down on equal protection grounds. Catch-22, again!
I hesitate to embrace this account, for three reasons. First, as I have discussed in this section, I have serious doubts about the whole enterprise of trying to uncover the motives of the Justices to discredit their decisions. Second, the Chief Justice and Justices Scalia and Thomas, all obviously pivotal in the Court's behavior throughout this saga, clearly would not have tolerated a recount of undervotes even if conducted under a uniform and entirely mechanical statewide standard. Their whole point was that no "undervotes" were "legal votes" under the Florida legislation that Article II made decisive. Bush, 121 S. Ct. at 537-38 (Rehnquist, C.J., concurring). They certainly cannot be accused, therefore, of laying a trap for the Florida court by brandishing the Article II sword on December 4. Third, the equal protection claims presented by the December 9 stay application that the Court treated as a certiorari petition and granted, were not mere retreads of the certiorari petitions that the Court had declined to hear on November 24. To be sure, the question of standardless recounts was posed on November 23, but not the question of counting undervotes and not overvotes, or the question of automatically including in the tally the partial recounts of four counties that included overvotes. More fundamentally, the November 23 petitions for certiorari challenged only inequalities in the protest phase of the election, before the certification of a result. The December 9 stay application challenged inequalities in the postcertification contest of the election results. Although I have vigorously disputed the coherence of drawing the line there, see supra section IV.B.3., pp. 258-63, the fact is that six Justices evidently disagreed, see Bush, 121 S. Ct. at 532; id. at 544 (Souter, J., concurring). If the view they held on December 12 was the same as on November 24, see Bush v. Palm Beach County Canvassing Bd., 121 S. Ct. 510 (2000) (mem.), there was no sandbagging in their failure to treat the equal protection challenge that they rejected with the same seriousness that they accorded the equal protection challenge that they later accepted.
n514. I explore the general question of motive in Laurence H. Tribe, The Mystery of Motive, Private and Public: Some Notes Inspired by the Problems of Hate Crime and Animal Sacrifice, 1993 Sup. Ct. Rev. 1 (1994).
n515. 347 U.S. 483 (1954).
n516. Dred Scott v. Sandford, 60 U.S. (19 How.) 393 (1857).
n517. See Dershowitz, supra note 2, at 12.
n518. See id. at 203-06.
n519. See supra pp. 289-90.
n520. Bickel, supra note 444, at 111.
n521. See, e.g., id. at 1-14.
n522. See supra pp. 277-79.
n523. Cf. Sanford Levinson, Constitutional Faith (1988).
n524. That such an extension, however desirable in itself, would link the fate of judicially prompted electoral reform to an ultimately rudderless vessel in Bush v. Gore is explained in section IV.C.1, pp. 268-70, above.
n525. See N.Y. Times Co., Inc., v. Tasini, 121 S. Ct. 2381 (2001); United States v. United Foods, Inc., 121 S. Ct. 2334 (2001).
n526. 478 U.S. 186 (1986).
n527. 500 U.S. 173 (1991). Don't get me wrong: I don't always think I should have won cases I lost. In hindsight, I do at times conclude that I may well have been wrong, and the Court right. See, e.g., Heffron v. Int'l Soc'y for Krishna Consciousness, Inc., 452 U.S. 640 (1981).
n528. 520 U.S. 351 (1997).
n529. This is, of course, a caricature of the most sophisticated defenses of judicial minimalism, see, e.g., Sunstein, supra note 413, but it is a widely held position to which such more sophisticated expositions might inadvertently contribute.
n530. See, e.g., Mark Tushnet, Taking the Constitution Away from the Courts (1999). To the degree that Professor Kramer's Foreword is understood to make a similar plea, I would respectfully dissent.
n531. 372 U.S. 335 (1963).
n532. 376 U.S. 254 (1964).
n533. 369 U.S. 186 (1962).