Judge Richard Posner of the United States Court of Appeals for the Seventh Circuit is an intellectual force to be reckoned with. The author, seemingly, of more books written while in active judicial service than many judges are of opinions, he can lay claim to the title of pre-eminent judicial theorist of our time. Nor are his opinions to be lightly dismissed as those of a right-leaning conservative in thrall to the Republican Party. His application of the precepts of the Chicago school of law and economics has led him, for instance, to endorse the right of gays to marry. More recently, he produced a review of the Lewinsky scandal in which he wisely found plenty of fault to go around for all the participants in the sorry spectacle–the President, the independent counsel, the leaders of the impeachment drive in Congress and, not least, the Supreme Court, for its naïve denial of executive privilege in Clinton v. Jones, which set the scene for much that followed. Judge Posner’s wide-ranging intellectual curiosity has produced such treasures as an eviscerating look at Janet Malcolm’s The Crime of Sheila McGough–a review for which Judge Posner read not only the book but also the transcript of the criminal trial that is the book’s subject.

On the subject of Bush v. Gore, Judge Posner’s efforts must be counted only partially successful. In the ten months since the Supreme Court settled the 2000 presidential election, the legal and academic community has weighed in with dozens of critical reviews. By and large, they haven’t been favorable. The epithets include lawless, posturing, disgraceful, illegitimate, unprincipled, outrageous, partisan, incomprehensible. But Posner has come up with a qualified defense of the Supreme Court’s actions in Breaking the Deadlock. Although he makes a fair case that the Florida Supreme Court manhandled the state’s election statutes, his effort to justify the US Supreme Court’s one-two punch–its December 9, 2000, stay of the Florida court’s order directing manual recounts and its December 12 decision to overturn that order at the expiration of the “safe harbor” period, with no remand for further proceedings–fails to persuade.

Judge Posner’s ultimate justification–that the Court saved us from ourselves or, more precisely, from the provisions for presidential elections prescribed by the Constitution and enabling acts of Congress–hints at an abandonment of the rule of law in the face of circumstances that were nowhere near as exigent as he suggests.

I.

In order to accord Judge Posner’s analysis the fair hearing it is due, we must begin with a recapitulation of the underlying legal events. Two rounds of litigation are at the heart of the story. The first of these originated under the Florida elections code as a protest proceeding brought by the Gore camp to demand a manual recount in four counties after Katherine Harris, the Secretary of State, had certified a victory for Governor Bush. The basis of Harris’s administrative action was a statutorily mandated automated recount, triggered by the neck-and-neck tally upon the close of the polls, that showed Governor Bush in the lead by 327 votes out of the nearly 6 million cast in the state.

On November 21, 2000, in a judicial-review proceeding brought in the trial court in Tallahassee to challenge the secretary’s ruling, the Florida Supreme Court held unanimously that a discrepancy between the machine count and a subsequent hand count amounted to an error in vote tabulation warranting a full manual recount and that the controlling Florida election statutes were in internal conflict on two critical issues: (1) the open-ended date for completion of a manual recount requested by a candidate under Florida statute § 102.166 could not be reconciled with the seven days after the election mandated under §§ 102.111 and 102.112 for submission of the completed count to state officials; and (2) the mandate in § 102.111 that the state electoral officials shall ignore late-filed county returns could not be reconciled with the permission in § 102.112 that they may do so. The Florida court concluded that in order to give effect to the county boards’ authority to undertake properly demanded manual recounts, the secretary must accept returns filed before the close of business on November 26–in effect, a twelve-day extension of the statutory deadline.

On December 4, 2000, the US Supreme Court unanimously vacated the Florida court’s judgment and remanded the case for clarification of the extent (if any) to which the state court had considered the impact of two provisions of federal law. The first of those provisions was the appointments clause of the Constitution, which sets out, in pertinent part, that each state shall appoint its presidential electors in such manner as the legislature thereof may direct. The second was a so-called safe-harbor statute, adopted as part of the Electoral Count Act after the Hayes-Tilden dispute in 1876 and codified as 3 USC § 5. That statute accords preclusive effect to a state’s appointment of presidential electors–so long as (1) they are appointed pursuant to a law in place before the date fixed for their appointment and (2) their identities are determined by resolution of any resultant electoral contest at least six days before the date prescribed for the vote of the Electoral College. Because the Florida legislature (like the legislature of every other state) had delegated to the voters of the state the selection of the presidential electors in a statute enacted well before November 7, 2000, and because the date fixed by federal law for the vote of the Electoral College was December 18, all hands agreed that if Florida were to avail itself of the safe-harbor provision, the state’s electors must be those chosen by the electorate on Election Day and identified by December 12. On December 11, 2000, the Florida Supreme Court filed its opinion following remand, giving consideration to the foregoing federal provisions and reinstating its prior determination.

The second round of litigation pertinent to Judge Posner’s analysis commenced after the Florida counties, with varying degrees of success in meeting the judicially fashioned deadline of November 26, certified their results–or, in the case of Miami-Dade County, which gave up on the manual recount midway through the process, their nonresults–to the Secretary of State. On November 26, following the expiration of the twelve-day grace period provided by the Florida Supreme Court in its first decision, Secretary Harris again certified Governor Bush as the winner, this time with a lead of 537 in the popular vote.

These administrative actions precipitated a contest proceeding under Florida statute §102.168, instituted as a civil action in the trial court in Tallahassee, brought by the Gore camp. After a two-day trial, Judge N. Sanders Sauls denied all relief, and the parties again repaired to the Florida Supreme Court. On December 8, the Florida high court, badly divided, reversed the trial court’s judgment. A bare majority of four justices held that the trial court had improperly deferred to the judgment of the administrative bodies that had conducted and overseen the recount; that the lower court had erred in refusing (1) to examine at all some 9,000 ballots that had registered no votes (so-called undervotes) in the presidential contest in Miami-Dade County when counted by machine and (2) to order that the election officials include in the final tally those votes for Gore and Bush, determined (1) by the Miami-Dade canvassing board before it gave up on its manual recount (with a net gain of 168 for Gore) and (2) by comparably determined votes in Palm Beach County (with a further net gain of 215 votes for Gore–or maybe 176, depending on whose count one accepted).

The Florida Supreme Court ordered an immediate resumption of the manual recount of the undervotes, to be supervised by the trial court in Tallahassee (before which disputed ballots had been transported), not just for Miami-Dade but throughout all counties in the state that had reported, but had not previously attempted to tabulate, undervoted ballots. The standard by which the manual counters were to ascertain the proper disposition of these undervotes, said the majority, was the “intent of the voter.”

On December 9, a Saturday, the US Supreme Court, on petition of the Bush camp, stayed the Florida Supreme Court’s judgment–and thereby the manual recount, then in progress–pending disposition of Governor Bush’s petition for review, which was granted and set for oral argument on the following Monday. The Supreme Court’s vote to stop the recount was 5 to 4.

On December 12, with only a few hours left before the expiration of the safe-harbor provided by 3 USC § 5, the Supreme Court rendered its final pronouncement in the case. Seven Justices–all but Justices Stevens and Ginsburg–thought that the absence of any standard imparted by the majority of the Florida high court to the manual recounters more informative than that they should “strive to ascertain the intent of the voter” implicated the equal protection clause or otherwise raised questions of fundamental fairness. Five of the seven Justices–excluding Justices Souter and Breyer–concluded that those constitutional concerns warranted reversal of the Florida court’s judgment ordering the manual recount. The five-Justice majority brooked no remand to the state court for another stab at a definition of the intent-of-the-voter standard, on the theory that the impending expiration of the safe-harbor period had put an end to the controversy. Three of the five Justices–Chief Justice Rehnquist and Justices Scalia and Thomas–would have held additionally that various aspects of the Florida court’s rulings concerning the interplay of the statutes making up the Florida electoral regime had violated the appointments clause. The blaze of dissents filed by Justices Stevens, Souter, Ginsburg and Breyer–each writing an opinion and joining variously in those of the other dissenters–suggested that at one point in the High Court’s deliberations, the Chief Justice’s opinion had commanded the support of a larger majority or had appeared imminently likely to do so.

This abbreviated summary does not address the litigation arising from unpostmarked absentee ballots from abroad (where, as the New York Times later demonstrated, different counties applied irreconcilably divergent standards), partisan assistance to would-be absentee voters on the part of some elections officials and the like; but the foregoing account suffices to set the stage for Judge Posner’s review.

II.

Judge Posner’s book is more than a lawyerly review of Bush v. Gore, although it certainly qualifies as that. In significant part, the book is a discourse on political philosophy, focused upon the role of general elections in the operation of a representative democracy. (They are a good thing, Judge Posner concludes, but not an end unto themselves.) The book is also a display of Judge Posner’s statistical erudition, replete with regression analyses of the Florida vote designed to demonstrate that factors such as voter error, illiteracy, income and race correlated more strongly with the rate of spoiled ballots in the 2000 election than did the factor of voting-machine failure.

Given the impact of antiquated punchcard mechanics on the outcome of the 2000 election, some observers might wonder whether the profound differences in the administration of Florida’s electoral system by the various counties around the state smacked of a federal constitutional problem. Some might even detect a hint of state-sponsored or -condoned discrimination, rising to the level of a denial of equal protection. That hint of illegality might ripen to a strong suspicion upon a showing that urban dwellers generally, blacks and Hispanics, and less well educated people suffered a diminution in the effectiveness of their vote because their ballots were disproportionately subject to disqualification in the punch-card voting process–precisely the showing that can be found in Judge Posner’s statistical analyses. Moreover, Posner recognizes, the disparities in voter accuracy attributable to divergent voting systems have long been known, though not well publicized.

The discovery that county-to-county differences in Florida’s voting systems resulted in real-world consequences leads Judge Posner ultimately to conclude that replacement of punch-card technology throughout the country would be a good policy prescription. But before he gets to that point, he explores, at least implicitly, the question of whether universal suffrage is a good idea. He concludes that it is, but only after considerable soul-searching about its consequences for the conduct of elections. “It would not be surprising,” he says in commenting on the makeup of the voters who cast spoiled ballots, “if a large fraction of the votes…had been cast by undecided, confused, clumsy, illiterate or semiliterate, or inexperienced (first time) voters.” Moreover, he surmises, the incompetent voters who spoiled their ballots in all likelihood were predominantly Democratic. His regression analyses persuade him that the punch-card ballot and county counting (rather than at the precinct level, where a watcher might catch spoiled ballots in time for voters to correct them), together with the correlated factors of low literacy, low income and being black, had a significant effect in increasing the frequency of spoiled votes. It may be all right to extend the franchise to people who cannot read, he concludes, but some conservatives may think it “rather an excess of democracy for illiterates to hold the electoral balance of power.” These are hardly the musings of a Jacksonian democrat.

Judge Posner acknowledges that “if punchcard technology had been replaced throughout Florida by marksense technology [i.e., by optical-scanning equipment], if all votes had been counted at the precinct level, if the butterfly ballot had not been used in Palm Beach County…and if the polling places had been better staffed and party activists had instructed their voters more carefully, it is quite likely that Gore would have won the popular vote in Florida on November 7 and thus would have become President without any recounting or litigation.” If just one of these fortuities had broken Gore’s way, he might be President. But if wishes were horses, beggars would ride.

Judge Posner adroitly harpoons the Florida Supreme Court, as have other commentators, on its extension of the seven-day limit on challenges to county certifications by a dozen days, from November 14 to November 26. Beyond that brazen act of judicial legislation, the Florida court’s action had the effect, no doubt unintended at the time, of foreshortening the period for the contest phase of the postelection proceeding–but only because the Florida court proceeded from the premise that the electoral controversy must be resolved by the safe-harbor date of December 12 in order to insure that Florida’s electoral votes would count in the Electoral College. That premise proved the undoing of the Florida court’s efforts–and of Gore’s candidacy–when the Supreme Court ultimately brought the election controversy to a close.

Judge Posner is no less kind–nor need he be–in addressing the Florida Supreme Court’s second decision, following the trial of the election contest before Judge Sauls. On the first round, the Florida high court had treated the county canvassing determinations challenged in the protest phase, leading to the Secretary of State’s certification of a winner, as of sufficient importance to justify, on equitable grounds, a twelve-day enlargement of the recount period unambiguously prescribed by the legislature. Now, the Florida court decreed, it was error for the trial court to have deferred to the Secretary of State’s administrative judgment at all. Instead, it would be necessary, upon remand, for the trial judge to oversee the manual counting of undervoted ballots–not just from those counties contested by Gore but from all over the state. Even though the proofs before Judge Sauls had shown that Miami-Dade and Palm Beach counties had employed variant and inconsistent standards in examining the ballots rejected by the machines, the trial judge on remand was given no guidance other than that he should discern the intent of the voter from the thousands of ballots to be examined by hand. To accomplish this heroic task while providing for judicial review within the time allotted before December 12, the drop-dead date so far as the federal safe harbor was concerned, plainly was impossible.

Most fundamentally, Judge Posner contends, the Florida Supreme Court erred at the outset, when it embarked on a re-examination of the automatic recount run by the machines. The justification for launching a countywide manual recount under the Florida statutory regime, we have seen, resides in an error in the tabulation of the votes–their counting, not their casting. A voter error in spoiling a ballot, rendering it uncounted by the machine, would not, on Judge Posner’s reading of the statute, constitute an error in tabulation. This is essentially the analysis, adopted early in the electoral dispute, by the Florida elections office overseen by Secretary Harris. In Judge Posner’s view, not only was this decision correct but the Florida courts should have deferred to it as an exercise of the administrator’s interpretation of the statutes whose workings she is empowered to supervise.

Because this is the linchpin of Judge Posner’s attack on the Florida Supreme Court, we should pause to consider its basis and implications. The governing statutory provision, found in Florida statute § 102.166(5), stipulates that, upon the losing candidate’s timely written request for a manual recount, the county canvassing board may undertake the recount (among three prescribed remedies, of which we shall have more to say in a moment) where a sample recount indicates an error in the vote tabulation that could affect the outcome of the election. Without doubt, there had been errors, and they could have affected the outcome, but were they errors in the “vote tabulation“? The phrase, notwithstanding Judge Posner, is not self-defining. It may, or may not, mean the same thing as the vote tabulation system–in other words, the machinery by which the ballots, once marked, are counted. The Florida court thought that the two phrases did not mean the same thing, and the court’s reasoning is not beyond the pale. Elsewhere in the very same section of the election code, the court noted, the legislature had used the phrase “vote tabulation system and automatic tabulating equipment.” Specifically, the legislature said, the county board could take remedial steps “if the [sample] recount indicate[d] an error in the vote tabulation.” The first of these was to recount all remaining precincts using the vote tabulation system; the second, to request that state officials verify the tabulation software; the third, to undertake a manual recount of all ballots. There would be no point in undertaking the sample manual recount if the only problem that warranted fixing was a problem with the machines–the board was empowered to look at the sample handcount to spot a difference from the machine count, whatever its source. This approach to the statute makes particularly good sense if, as Judge Posner says, problems with the machine count of punch-card ballots are a known factor but the cost of replacing punch-card systems is prohibitive.

It is an accepted canon of statutory interpretation, with which Judge Posner surely is familiar, that when the legislature uses one phrase in one place in a statute and another, different phrase in another place in the same statute, each phrase should be accorded meaning, and it cannot be assumed that the two phrases mean the same thing, as though the legislature had used the same words twice. There is no presumption that legislators are sloppy when they write statutes, any more than that judges are sloppy when they write opinions.

This does not mean that Judge Posner’s interpretation is wrong. He might be right, and his reasoning might be better than that of the Florida Supreme Court. But it seems wrong of him to conclude that the Florida court was so obviously off-base that the US Supreme Court was within bounds to reject the Florida court’s reading of Florida law. And it is doubly wrong to treat the question of interpretation of the Florida statute as a de novo question, rather than as one subject to some level of deference when the question came before the US Supreme Court–a point to which we shall revert when we consider Judge Posner’s treatment of the Supreme Court’s decision on the second and final round.

A statutory phrase is to be read not in isolation, as though it were a snippet justifiably torn from its context, but with reference to its policy objectives. Often those objectives are illuminated by companion provisions in the same statute. In the Florida election code, just two paragraphs down from the tabulation provision, appears a directive to the canvassing boards that in conducting their manual recounts, they are to strive to ascertain the intent of the voters (Florida statute § 102.166[7]). That language provides a further basis to believe that the legislature chose not to rest the electoral process entirely with machines but to provide a safety valve of manual recounting in aid of discerning the will of the electorate where the boards found an error in the tabulation.

Further evidence of the legislative intent is found in Florida statute § 101.5614, an omnibus provision that instructs the canvassing boards how to deal with spoiled votes of every kind–electronic or electromechanical, ballot cards, ballots with and without the office up for election printed directly on them. In the case of ballot cards damaged or defective so that they cannot properly be counted by the automatic tabulating equipment, subdivision (5) specifies: “No [such] 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.” That provision surely looks like a prescription for remedial measures where the counting machinery, for whatever reason, has failed to tabulate a ballot for which the voter’s intent is clear.

That is not the lesson drawn by Judge Posner. To the contrary, he argues,

To classify a “failure” that is built into the design of the tabulating machinery [such as the failure to count punchcard ballots not properly cleared of chads] as an error or defect in the tabulation of the vote would make hand recounts mandatory throughout most of the state in all close elections–something the election statute cannot reasonably be read to contemplate.

As if to insure that the point is not lost, Judge Posner reiterates: “Hand recounts in close elections would become the rule rather than the exception. This is all upside down.”

These remarks mirror the observation of Chief Justice Rehnquist, in his concurring opinion in the second and final round before the Supreme Court, that it would be “absurd” to attribute to the Florida legislature the design of counting by hand when the election is close because the machines could be counted upon to miss lawful ballots. This is strange thinking. Hand recounts, however unattractive, might be seen as necessary to insure fidelity to the popular will–the position that seems to be what the legislature spelled out in the statutes governing the count.

Attributing primary significance to capturing every ballot where the will of the voter may be discerned presupposes that elections have an importance unto themselves. Judge Posner, as political philosopher, tells us that we should not think of elections in that fashion. They are but a means to an end, rather than an end in themselves: “Elections are part of the framework and so belong to the domain of formal justice,” says Judge Posner. “They are not actualizations of the popular will.” This is a rather bloodless way of viewing the process that imparts legitimacy to elected officials and, through them, to our representative form of government.

Whatever the weaknesses in his argument, Judge Posner is far better at demolishing the work of the Florida court than he is at upholding that of the US Supreme Court. He acknowledges that the equal-protection argument that persuaded a bare majority of five Justices (and possibly two others besides, who voiced concern about the unfairness of the standardless recount and joined in the per curiam opinion announcing the reversal of the Florida court’s order) was not a persuasive ground. He would have been hard-pressed to come to a different conclusion. For starters, Governor Bush had no standing to complain of an equal-protection violation: He was not a Florida voter, but at best the beneficiary of those Florida citizens who had voted for him. (Judge Posner does not mention this point, although he is quick to point out that Gore lacked standing in the lawsuit that his camp brought.) The standing of the citizens who had voted for Bush was better, but not by much. Those of them who might have attempted to cast a ballot for Bush but who had spoiled it by failing to detach the chads hardly could be heard to complain that their votes now might be counted. Those who had marked their ballots properly might fear the dilution of their voting strength, however infinitesimally, by the inclusion of spoiled ballots on the manual recount, but there was no way of knowing ex ante which candidate would have benefited from that recount. For all that appeared at the time, and for all we know today, a manual recount would have solidified Bush’s position as the leader in the popular vote.

More fundamentally, there was no suggestion that a recount based on the “intent of the voter” standard would work a systemic, deliberate discrimination against a discrete, insular minority, whether blacks (as in poll-tax cases), illiterates (as in literacy cases) or urban dwellers (as in reapportionment cases). One might well ask: Where was the invidiousness that is the hallmark of a denial of equal protection? There was none, as Judge Posner notes when he considers the disparate impact of the use of punch-card technology and accordingly passes over the suggestion that the voting regime itself worked a large-scale denial of equal protection, as against those voters whose spoiled ballots the counting machines did not tally.

The difference in voting accuracy achieved by the various counties in Florida might well have been deemed a denial of equal protection. In that sense, the Florida voting system, as applied, was not “an innocent law that just happen[ed] to have an unequal impact.” Judge Posner does not address this possibility.

A better argument, Judge Posner thinks, is the one that drew only three Justices–that the Florida Supreme Court had so altered the legislative scheme enacted by the Florida legislature for the selection of the state’s electors that the lower court had violated the appointments clause (the mandate of the federal Constitution that the presidential electors be chosen in the manner directed by the legislature). Strangely, Judge Posner does not delve into this argument much, although he makes plain that he would have favored a reversal of the lower court’s decision on appointments-clause grounds rather than equal-protection grounds if those had been the only options.

We need not resort to extravagant hypotheticals to test the appointments-clause argument. We have seen that Judge Posner rests his criticism of the Florida Supreme Court–not exclusively, by any means, but to a considerable degree–on that court’s interpretation of the word “tabulation” in one section of the Florida election statutes. We have also seen that the Florida court’s interpretation of this word was not so far-fetched as to warrant the conclusion that the court was wrong, not even on de novo review. But de novo was the standard that the three concurring Justices of the US Supreme Court seemed to use. They decided for themselves what the Florida statute meant, without any intimation that they had deferred to the Florida court’s reading or that they ought to do so. Yet, ordinarily, federal courts yield to a state court’s determination of what a state statute means. Perhaps a state court could not be allowed to say that black means white, but surely it could say that salmon means pink. The Florida court’s reading of the “tabulation” phrase looks a lot more like salmon/pink than black/white.

The argument in support of the US Supreme Court’s decision that most attracts Judge Posner is one never made by the Court at all–and for obvious reasons. The argument, at its nub, is that if the Court had not acted on December 12 to cut off the election dispute, the controversy could have spilled over into the New Year, with resultant chaos. This argument, which we may term the Louis XIV thesis (après moi, le déluge), is portrayed in a parade of horribles: two competing electoral slates from Florida; Congressional and judicial challenges to one or both of them; a political uproar in the House; a Senate evenly divided between the Republicans and Democrats; and the prospect of an acting President–Dennis Hastert (who might decline the honor rather than give up the House Speakership), Strom Thurmond (who might well be oblivious of his surroundings), Madeleine Albright (probably ineligible because not native-born), or Lawrence Summers, as holdover Secretary of the Treasury (our first Jewish President, pace Joe Lieberman).

This conjures an impressive parade of improbabilities, but a break in the chain at any point would have brought the parade to an end. The mechanics of the manual recount likely would have caused the process to collapse in a matter of days. If the contest had gone forward, the Republican-controlled legislature in Florida would have nominated a slate of electors. The Governor (Bush’s brother) and the Secretary of State (who had participated actively in his campaign) would have lent their approval, fulminations of the Florida courts to the contrary notwithstanding. In so doing, the legislature would have given up the federal safe harbor, because the legislators would have selected the presidential electors under the aegis of a law (or resolution) enacted after the general election; but that would mean only that under the Electoral Count Act, somebody in Congress might challenge the Florida electors when it came time to open the envelopes. That challenge would require the concurrence of a majority of both the House and the Senate, with the former body controlled by the Republicans and the latter evenly divided between the parties.

All this presupposes a vote count trending in Gore’s favor. Given the closeness of the vote as tabulated and the uncertainty of the spoiled ballots that remained, it is of roughly equal possibility that Bush would have continued to lead in the recount by the time the dispute reached Congress. In that case, the Florida legislature would have been all the more likely to act in his favor and the Congress in Washington to have acquiesced. And Bush’s election would have been buttressed by the widespread acceptance that he had won the vote in Florida, counted perhaps with less than perfect guarantees of accuracy and under trying circumstances, but counted.

Letting the electoral dispute go forward without judicial intervention would have had the advantage of giving effect to the Constitution–a goal that we like to think is favored by the Supreme Court. However rough and tumble the political process might be, the election of a President should be a political event, not the subject of judicial decision. If it is permissible to justify the Supreme Court’s action on the ground that the five-Justice majority acted pragmatically, albeit without legal justification, to save the country from possible political turmoil, it should be equally permissible to justify a renegade Army general surrounding the Capitol with troops to enforce a presidential selection, on just the same reasoning.

It is the political nature of the controversy, as well as the untested provisions of the Electoral Count Act and related statutes, that lead Judge Posner to surmise that if the contending factions had petitioned the Supreme Court for review during this cavalcade of tumultuous events, the Court likely would have turned such requests aside on grounds that they presented a political question. But if the Court properly could have left the country high and dry in January 2001, it is hard to understand why the Court would not have done even better to forswear its involvement in December 2000, when so many additional options were available short of judicial intervention. Judge Posner’s appeal to pragmatism as a justifiable, albeit unarticulated ground of decision seems less pragmatic than he would have us believe.

Posner’s preference for judicial resolution is a sign of judicial elitism–a call for more judicial intervention, not less. His logic cannot be confined to the electoral process (where the prospect of continual judicial intervention is bad enough), but is sufficiently expansive to accommodate just about everything. If judges should take the heat off legislators in resolving electoral contests, perhaps they should do likewise in resolving other political hot potatoes as well.

The scariest of Judge Posner’s deluge scenarios is that the controversy over the Florida electors might have forced the election of the President into the House of Representatives and of the Vice President into the Senate, neither Bush nor Gore having achieved a majority of the electors appointed by the several states. That scenario is not only anticipated in the Constitution; it describes a circumstance that the drafters of the Constitution expected would be more or less routine. The presidential electors, scattered among the several states, were directed to vote in their respective state capitals all on the same day. They were not to travel to the national capital to cast their ballots, nor were they to indulge in repeated ballots, like the political conventions or the papal elections, until a majority candidate emerged the winner. These restrictions provided a safeguard against coercion and bribery. The electors were to cast only one round of ballots, which was thought to be unlikely to produce a majority for any one candidate. The election then would go to the House, there to be decided with each state having one vote.

The constitutional process might be messy, and it might yield a late or unsatisfactory result, but it has the advantage of being the process prescribed by the Constitution. The Supreme Court Justices took an oath to uphold the Constitution, not to decide that its prescriptions are too risky. Judge Posner’s prescription for the pragmatic point of view turns out to be profoundly undemocratic, and unconstitutional besides. Because the Court saved us from ourselves, he says, the decision passes muster, even though, he acknowledges, “there are respectable schools of jurisprudence according to which Bush v. Gore could be shown to be unprincipled, even usurpative.” In the final analysis, this is a damning indictment of the decision that goes far to undermine Judge Posner’s conclusion that the Court acted properly in pragmatic fashion.

III.

Whatever the rights and wrongs of the Supreme Court’s decision on the merits, there remain the issues posed by the Court’s choice of remedies. Here Judge Posner is even more elliptical, and even less convincing, than in his defense of the Court’s decision on the merits.

The remedies problem has two components–the stay of the manual recount ordered by five Justices on Saturday, December 9, and the judgment of reversal without remand, ordered by the same five Justices three days later, on Tuesday, December 12, at 10 pm Eastern time, two hours shy of the expiration of the safe-harbor period. If the Florida Supreme Court had erred in ordering a manual recount without providing adequate standards to the counters, the logical remedy would have been to reverse and remand with instructions to come up with a better standard. This was the point urged in dissent by Justices Souter and Breyer, who otherwise agreed with the five Justices in the majority that the absence of definition in the “intent of the voter” test had constitutional implications.

At the end of its opinion, the five-Justice majority dispensed with the notion that the case should be remanded for further proceedings. There was no time left for standards, the majority said; the safe harbor was about to expire. Judge Posner correctly notes the “gotcha” flavor to this argument. The Florida Supreme Court had intimated that the electoral dispute should be brought to an end by December 12 in order to insure that Florida’s electors would vote in the Electoral College on December 18. But the Florida court had never said that such was the law in Florida. Certainly the legislature had never said so. Who knows what the court or the legislature might have said when confronted, on the one hand, with a looming deadline for the safe harbor and, on the other, with an imperative to count all the votes where the intent of the voter fairly could be discerned?

There is something ironic about the disposition of a case, founded in major part upon the criticism that the lower court had invented a law never enacted by its state’s legislature, that rests for its remedy upon, well, the invention of a law never enacted by the legislature.

Judge Posner is dismissive of objections to the stay entered by the Court on December 9, the Saturday surprise that put paid to the manual recount and to Gore’s chances. Because the Court stopped the recount for good only three days later, Judge Posner argues, the stay didn’t matter much, and further, balancing the risk of harm to Bush with his case on the merits, the grant of the stay was not out of line. Judge Posner is wrong on both counts.

The stay mattered a great deal. It stopped the manual recount dead in its tracks while Bush led in the tally, and it precluded any likelihood that Gore might overtake him. If, as Judge Posner suggests, the five Justices who voted for the stay already had decided that Bush had a convincing case on the merits, the only point of the stay was to insulate their decision on the merits from whatever additional criticism would have flowed from Gore’s having taken the lead in the recount in the interim between the Florida Supreme Court’s order directing the manual recount and the Supreme Court’s ultimate disposition.

The grant of a stay pending appeal, akin to the grant of a preliminary injunction in the trial court, is an extraordinary event. It presupposes some combination of likelihood of success on the merits, irreparable injury in the interim, a balance of advantage weighing in favor of the applicant and no adverse impact on the public interest. Bush’s application for a stay met none of these criteria. On Judge Posner’s own reasoning, the equal-protection argument was not meritorious, and the appointments-clause argument ultimately corralled only three votes. Whatever detriment Bush might have suffered if the recount continued was counterbalanced by the detriment to Gore if the recount was halted. The public surely had an interest in knowing the disposition of the spoiled ballots, rejected by the counting machines but readable by human beings.

Lacking in Bush’s plea for interim injunctive relief was any sign of irreparable injury. Irreparable injury presupposes the unavailability of an adequate remedy at law. Bush had several adequate remedies at law–the ones prescribed under Florida statute, the Electoral Count Act and the Constitution. If he did not care for the results of the recount or the manner in which the recount was conducted, he could contest the inclusion of the votes before the Florida courts or in a fresh suit instituted in federal court. (He already had done that once, notwithstanding his supposed preference for state rather than federal resolution of state issues.) He could have petitioned the Florida legislature to appoint electors favorable to him (which the legislature surely would have done). He could have used the Republican Party’s control of the House to block any effort to accept presidential electors from Florida pledged to Gore rather than to Bush. To the extent that the manual recount was a mess, his arguments before each of these forums would have looked pretty good.

Judge Posner does not touch upon the availability of an adequate remedy at law, perhaps because his chaos scenario rules out the notion that the panoply of legal remedies available to Bush was adequate to the occasion. Nor does Judge Posner have much to say for the majority’s rationale in imposing the stay–a one-paragraph per curiam that announced the fact of the stay without an iota of explanation. This is a surprising lapse, given that the lower federal judges, to whom Bush had presented his equal-protection arguments, had declined to consider them precisely on the ground that Bush’s injury was not irreparable.

It was left to Justice Scalia, speaking for himself, to offer a justification, which Judge Posner mildly critiques as possibly not the best work of which the Justice was capable but which Posner does not otherwise discuss. At the risk of prolonging this already overlong review, we should tarry a moment. Here is the operative passage from Justice Scalia’s concurrence in the stay, with his points numbered for ease of reference:

[1] The counting of votes that are of questionable legality does in my view threaten irreparable harm to petitioner [i.e., to Bush], and to the country, by casting a cloud upon what he claims to be the legitimacy of his election. [2] Count first, and rule upon legality afterwards, is not a recipe for producing election results that have the public acceptance democratic stability requires. [3] Another issue in the case, moreover, is the propriety, indeed the constitutionality, of letting the standard for determination of voters’ intent–dimpled chads, hanging chads, etc.–vary from county to county, as the Florida Supreme Court opinion, as interpreted by the Circuit Court, permits. [4] If petitioner is correct that counting in this fashion is unlawful, permitting the count to proceed on that erroneous basis will prevent an accurate recount from being conducted on a proper basis later, since it is generally agreed that each manual recount produces a degradation of the ballots, which renders a subsequent recount inaccurate.

Not one of these arguments holds water. (1) If the spoiled ballots were illegal and if Florida law ultimately were held to bar their inclusion, a court could order that they not be counted. That is just what the Supreme Court did three days after the stay. In any event, Bush had no legally cognizable right to be free of clouds; he had only whatever legal right flowed from Secretary Harris’s certification that he was the winner after the delayed recount on the protest phase ended on November 26. That certificate, without more, did not entitle him to an injunction of a proceeding, otherwise lawful, that might show the certificate to have been issued in error.

(2) Counting first and then setting the rules was an effective argument only if one assumed that no rules at all would be in place during and after the count. That was a doubtful proposition, given the presence of a trial-court judge overseeing the process in Tallahassee on remand from the Florida Supreme Court, and the availability of state and federal appellate review thereafter.

(3) The manual recount was not to be subject to the vagaries of county boards, although there had been vagaries aplenty in the manual recounts attempted to date. The Florida Supreme Court’s order remanding the contest to Judge Lewis, with directions to oversee a recount, provided that the “intent of the voter” standard was to be applied, ultimately, by a single trial judge sitting in Tallahassee. That judge would have been capable of responding to an argument from Bush that different counties had counted spoiled ballots in different ways and of doing something about it.

(4) If the act of recounting destroyed the evidence, that destruction would be one more arrow in the quiver of whichever candidate–Gore or Bush–decided to challenge the outcome of the recount in the various venues provided under Florida and federal law. But the risk of alteration of the ballots provided no basis to halt the recount. Gore, as the party pressing for the recount, had taken his chances that a court ultimately might determine the process too rushed and uncertain to warrant judicial confidence, and then he would be out of luck if alteration of the evidence foreclosed a further manual recount.

So the stay was improvident at best, and reckless at worst. Judge Posner’s diffident defense of it is a signal that the remedial aspect of the Supreme Court’s handiwork was the least justifiable aspect of an already shaky product.

IV.

Some commentators think the Supreme Court’s decision in Bush v. Gore ranks as the worst ever. Given the contenders for that crown, it’s a tough call. As Alexander Bickel once famously said, with an active and ongoing institution, it’s unwise to speak of rock bottom. Judge Posner is inclined to be more charitable: not the greatest reasoning in the world, perhaps, but better than a poke in the eye with a sharp stick.

To blame for the fuss, Posner thinks, is everybody but the five Justices in the majority: the four dissenters, who fouled their own nest by attacking the majority’s reasoning; Gore, who commenced the legal proceedings (although Bush, he notes, was the first to file suit, and in a federal court at that); and a liberal-left law-school professoriate, whose self-serving policy preferences are exceeded, in Posner’s view, only by its collective ignorance of constitutional law. The people, he is confident, will take the long view, if they haven’t forgotten the case already. (Posner was writing before Bush became a wartime President.)

Judge Posner is not overly concerned that the Supreme Court’s reputation will suffer as a consequence of its decision. In truth, the Supreme Court left no winners on the field. Vice President Gore lost, to a certainty. Governor Bush lost (even though he won), because the halt in the recount procured at his behest foreclosed any possibility that the recount might have proved him the winner (as several postelection reconstructions have suggested he would have been) and undermined his claim to be the legitimately as well as the legally elected President. The Court has done itself inestimable damage among the citizens of this country, who may be more interested in these matters than Judge Posner believes and who now will think twice before they regard the Supreme Court Justices as different in kind, rather than in degree, from self-acknowledged politicians.

Judge Posner has contributed a valuable and even indispensable work to the debate over the 2000 election. However, his book is unlikely to bring comfort to the reader that justice was done, either in the confines of judicial decision-making or in the larger spaces of presidential politics.