SUPREME COURT OF THE UNITED STATES

-----------------

No. 00-949

-----------------

GEORGE W. BUSH, ET AL., PETITIONERS v. ALBERT GORE, JR., ET AL.

ON WRIT OF CERTIORARI TO THE FLORIDA SUPREME COURT

[December 12, 2000]

JUSTICE GINSBURG, with whom JUSTICE STEVENS joins, and with whom JUSTICE SOUTER and JUSTICE BREYER join as to Part I, dissenting.

I

The CHIEF JUSTICE acknowledges that provisions of Florida's Election Code "may well admit of more than one interpretation." Ante, at 3. But instead of respecting the state high court's province to say what the State's Election Code means, THE CHIEF JUSTICE maintains that Florida's Supreme Court has veered so far from the ordinary practice of judicial review that what it did cannot properly be called judging. My colleagues have offered a reasonable construction of Florida's law. Their construction coincides with the view of one of Florida's seven Supreme Court justices. Gore v. Harris, __ So. 2d __, __ (Fla. 2000) (slip op., at 45-55) (Wells, C. J., dissenting); Palm Beach County Canvassing Bd. v. Harris, __ So. 2d __, __ (Fla. 2000) (slip op., at 34) (on remand) (confirming, 6-1, the construction of Florida law advanced in Gore). I might join THE CHIEF JUSTICE were it my commission to interpret Florida law. But disagreement with the Florida court's interpretation of its own State's law does not warrant the conclusion that the justices of that court have legislated. There is no cause here to believe that the members of Florida's high court have done less than "their mortal best to discharge their oath of office," Sumner v. Mata, 449 U. S. 539, 549 (1981), and no cause to upset their reasoned interpretation of Florida law.

This Court more than occasionally affirms statutory, and even constitutional, interpretations with which it disagrees. For example, when reviewing challenges to administrative agencies' interpretations of laws they implement, we defer to the agencies unless their interpretation violates "the unambiguously expressed intent of Congress." Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837, 843 (1984). We do so in the face of the declaration in Article I of the United States Constitution that "All legislative Powers herein granted shall be vested in a Congress of the United States." Surely the Constitution does not call upon us to pay more respect to a federal administrative agency's construction of federal law than to a state high court's interpretation of its own state's law. And not uncommonly, we let stand state-court interpretations of federal law with which we might disagree. Notably, in the habeas context, the Court adheres to the view that "there is 'no intrinsic reason why the fact that a man is a federal judge should make him more competent, or conscientious, or learned with respect to [federal law] than his neighbor in the state courthouse.'" Stone v. Powell, 428 U. S. 465, 494, n. 35 (1976) (quoting Bator, Finality in Criminal Law and Federal Habeas Corpus For State Prisoners, 76 Harv. L. Rev. 441, 509 (1963)); see O'Dell v. Netherland, 521 U. S. 151, 156 (1997) ("[T]he Teague doctrine validates reasonable, good-faith interpretations of existing precedents made by state courts even though they are shown to be contrary to later decisions." (citing Butler v. McKellar, 494 U. S. 407, 414 (1990)); O'Connor, Trends in the Relationship Between the Federal and State Courts from the Perspective of a State Court Judge, 22 Wm. & Mary L. Rev. 801, 813 (1981) ("There is no reason to assume that state court judges cannot and will not provide a 'hospitable forum' in litigating federal constitutional questions.").

No doubt there are cases in which the proper application of federal law may hinge on interpretations of state law. Unavoidably, this Court must sometimes examine state law in order to protect federal rights. But we have dealt with such cases ever mindful of the full measure of respect we owe to interpretations of state law by a State's highest court. In the Contract Clause case, General Motors Corp. v. Romein, 503 U. S. 181 (1992), for example, we said that although "ultimately we are bound to decide for ourselves whether a contract was made," the Court "accord[s] respectful consideration and great weight to the views of the State's highest court." Id., at 187 (citation omitted). And in Central Union Telephone Co. v. Edwardsville, 269 U. S. 190 (1925), we upheld the Illinois Supreme Court's interpretation of a state waiver rule, even though that interpretation resulted in the forfeiture of federal constitutional rights. Refusing to supplant Illinois law with a federal definition of waiver, we explained that the state court's declaration "should bind us unless so unfair or unreasonable in its application to those asserting a federal right as to obstruct it." Id., at 195.1

In deferring to state courts on matters of state law, we appropriately recognize that this Court acts as an " 'outside[r]'lacking the common exposure to local law which comes from sitting in the jurisdiction." Lehman Brothers v. Schein, 416 U. S. 386, 391 (1974). That recognition has sometimes prompted us to resolve doubts about the meaning of state law by certifying issues to a State's highest court, even when federal rights are at stake. Cf. Arizonans for Official English v. Arizona, 520 U. S. 43, 79 (1997) ("Warnings against premature adjudication of constitutional questions bear heightened attention when a federal court is asked to invalidate a State's law, for the federal tribunal risks friction-generating error when it endeavors to construe a novel state Act not yet reviewed by the State's highest court."). Notwithstanding our authority to decide issues of state law underlying federal claims, we have used the certification devise to afford state high courts an opportunity to inform us on matters of their own State's law because such restraint "helps build a cooperative judicial federalism." Lehman Brothers, 416 U. S., at 391.

Just last Term, in Fiore v. White, 528 U. S. 23 (1999), we took advantage of Pennsylvania's certification procedure. In that case, a state prisoner brought a federal habeas action claiming that the State had failed to prove an essential element of his charged offense in violation of the Due Process Clause. Id., at 25-26. Instead of resolving the state-law question on which the federal claim depended, we certified the question to the Pennsylvania Supreme Court for that court to "help determine the proper state-law predicate for our determination of the federal constitutional questions raised." Id., at 29; id., at 28 (asking the Pennsylvania Supreme Court whether its recent interpretation of the statute under which Fiore was convicted "was always the statute's meaning, even at the time of Fiore's trial"). THE CHIEF JUSTICE's willingness to reverse the Florida Supreme Court's interpretation of Florida law in this case is at least in tension with our reluctance in Fiore even to interpret Pennsylvania law before seeking instruction from the Pennsylvania Supreme Court. I would have thought the "cautious approach" we counsel when federal courts address matters of state law, Arizonans, 520 U. S., at 77, and our commitment to "build[ing] cooperative judicial federalism," Lehman Brothers, 416 U. S., at 391, demanded greater restraint.

Rarely has this Court rejected outright an interpretation of state law by a state high court. Fairfax's Devisee v. Hunter's Lessee, 7 Cranch 603 (1813), NAACP v. Alabama ex rel. Patterson, 357 U. S. 449 (1958), and Bouie v. City of Columbia, 378 U. S. 347 (1964), cited by THE CHIEF JUSTICE, are three such rare instances. See ante, at 4, 5, and n. 2. But those cases are embedded in historical contexts hardly comparable to the situation here. Fairfax's Devisee, which held that the Virginia Court of Appeals had misconstrued its own forfeiture laws to deprive a British subject of lands secured to him by federal treaties, occurred amidst vociferous States'rights attacks on the Marshall Court. G. Gunther & K. Sullivan, Constitutional Law 61-62 (13th ed. 1997). The Virginia court refused to obey this Court's Fairfax's Devisee mandate to enter judgment for the British subject's successor in interest. That refusal led to the Court's pathmarking decision in Martin v. Hunter's Lessee, 1 Wheat. 304 (1816). Patterson, a case decided three months after Cooper v. Aaron, 358 U. S. 1 (1958), in the face of Southern resistance to the civil rights movement, held that the Alabama Supreme Court had irregularly applied its own procedural rules to deny review of a contempt order against the NAACP arising from its refusal to disclose membership lists. We said that "our jurisdiction is not defeated if the nonfederal ground relied on by the state court is without any fair or substantial support." 357 U. S., at 455. Bouie, stemming from a lunch counter "sit-in" at the height of the civil rights movement, held that the South Carolina Supreme Court's construction of its trespass laws -- criminalizing conduct not covered by the text of an otherwise clear statute -- was "unforeseeable" and thus violated due process when applied retroactively to the petitioners. 378 U. S., at 350, 354.

THE CHIEF JUSTICE's casual citation of these cases might lead one to believe they are part of a larger collection of cases in which we said that the Constitution impelled us to train a skeptical eye on a state court's portrayal of state law. But one would be hard pressed, I think, to find additional cases that fit the mold. As JUSTICE BREYER convincingly explains, see post, at 5-9 (dissenting opinion), this case involves nothing close to the kind of recalcitrance by a state high court that warrants extraordinary action by this Court. The Florida Supreme Court concluded that counting every legal vote was the overriding concern of the Florida Legislature when it enacted the State's Election Code. The court surely should not be bracketed with state high courts of the Jim Crow South.

THE CHIEF JUSTICE says that Article II, by providing that state legislatures shall direct the manner of appointing electors, authorizes federal superintendence over the relationship between state courts and state legislatures, and licenses a departure from the usual deference we give to state court interpretations of state law. Ante, at 5 ("To attach definitive weight to the pronouncement of a state court, when the very question at issue is whether the court has actually departed from the statutory meaning, would be to abdicate our responsibility to enforce the explicit requirements of Article II."). The Framers of our Constitution, however, understood that in a republican government, the judiciary would construe the legislature's enactments. See U. S. Const., Art. III; The Federalist No. 78 (A. Hamilton). In light of the constitutional guarantee to States of a "Republican Form of Government," U. S. Const., Art. IV, ¤4, Article II can hardly be read to invite this Court to disrupt a State's republican regime. Yet THE CHIEF JUSTICE today would reach out to do just that. 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. See, e.g., Gregory v. Ashcroft, 501 U. S. 452, 460 (1991) ("Through the structure of its government, and the character of those who exercise government authority, a State defines itself as a sovereign."); Highland Farms Dairy, Inc. v. Agnew, 300 U. S. 608, 612 (1937) ("How power shall be distributed by a state among its governmental organs is commonly, if not always, a question for the state itself.").2 Article II does not call for the scrutiny undertaken by this Court.

The extraordinary setting of this case has obscured the ordinary principle that dictates its proper resolution: Federal courts defer to state high courts'interpretations of their state's own law. This principle reflects the core of federalism, on which all agree. "The Framers split the atom of sovereignty. It was the genius of their idea that our citizens would have two political capacities, one state and one federal, each protected from incursion by the other." Saenz v. Roe, 526 U. S. 489, 504, n. 17 (1999) (citing U. S. Term Limits, Inc. v. Thornton, 514 U. S. 779, 838 (1995) (KENNEDY, J., concurring)). THE CHIEF JUSTICE's solicitude for the Florida Legislature comes at the expense of the more fundamental solicitude we owe to the legislature's sovereign. U. S. Const., Art. II, ¤1, cl. 2 ("Each State shall appoint, in such Manner as the Legislature thereof may direct," the electors for President and Vice President) (emphasis added); ante, at 1-2 (STEVENS, J., dissenting).3 Were the other members of this Court as mindful as they generally are of our system of dual sovereignty, they would affirm the judgment of the Florida Supreme Court.

II

I agree with JUSTICE STEVENS that petitioners have not presented a substantial equal protection claim. Ideally, perfection would be the appropriate standard for judging the recount. But we live in an imperfect world, one in which thousands of votes have not been counted. I cannot agree that the recount adopted by the Florida court, flawed as it may be, would yield a result any less fair or precise than the certification that preceded that recount. See, e.g., McDonald v. Board of Election Comm'rs of Chicago, 394 U.S. 802, 807 (1969) (even in the context of the right to vote, the state is permitted to reform " 'one step at a time'") (quoting Williamson v. Lee Optical of Oklahoma, Inc., 348 U.S. 483, 489 (1955)).

Even if there were an equal protection violation, I would agree with JUSTICE STEVENS, JUSTICE SOUTER, and JUSTICE BREYER that the Court's concern about "the December 12 deadline," ante, at 12, is misplaced. Time is short in part because of the Court's entry of a stay on December 9, several hours after an able circuit judge in Leon County had begun to superintend the recount process. More fundamentally, the Court's reluctance to let the recount go forward -- despite its suggestion that "[t]he search for intent can be confined by specific rules designed to ensure uniform treatment,"ante, at 8 -- ultimately turns on its own judgment about the practical realities of implementing a recount, not the judgment of those much closer to the process.

Equally important, as JUSTICE BREYER explains, post, at 12 (dissenting opinion), the December 12 "deadline" for bringing Florida's electoral votes into 3 U. S. C. ¤5's safe harbor lacks the significance the Court assigns it. Were that date to pass, Florida would still be entitled to deliver electoral votes Congress must count unless both Houses find that the votes "ha[d] not been . . . regularly given." 3 U. S. C. ¤15. The statute identifies other significant dates. See, e.g., ¤7 (specifying December 18 as the date electors "shall meet and give their votes"); ¤12 specifying "the fourth Wednesday in December" -- this year, December 27 -- as the date on which Congress, if it has not received a State's electoral votes, shall request the state secretary of state to send a certified return immediately). But none of these dates has ultimate significance in light of Congress' detailed provisions for determining, on "the sixth day of January,"the validity of electoral votes. ¤15.

The Court assumes that time will not permit "orderly judicial review of any disputed matters that might arise." Ante, at 12. But no one has doubted the good faith and diligence with which Florida election officials, attorneys for all sides of this controversy, and the courts of law have performed their duties. Notably, the Florida Supreme Court has produced two substantial opinions within 29 hours of oral argument. In sum, the Court's conclusion that a constitutionally adequate recount is impractical is a prophecy the Court's own judgment will not allow to be tested. Such an untested prophecy should not decide the Presidency of the United States.

I dissent.

1 See also Lucas v. South Carolina Coastal Council, 505 U. S. 1003, 1032, n. 18 (1992) (South Carolina could defend a regulatory taking "if an objectively reasonable application of relevant precedents [by its courts] would exclude . . . beneficial uses in the circumstances in which the land is presently found"); Bishop v. Wood, 426 U. S. 341, 344-345 (1976) (deciding whether North Carolina had created a property interest cognizable under the Due Process Clause by reference to state law as interpreted by the North Carolina Supreme Court). Similarly, in Gurley v. Rhoden, 421 U. S. 200 (1975), a gasoline retailer claimed that due process entitled him to deduct a state gasoline excise tax in computing the amount of his sales subject to a state sales tax, on the grounds that the legal incidence of the excise tax fell on his customers and that he acted merely as a collector of the tax. The Mississippi Supreme Court held that the legal incidence of the excise tax fell on petitioner. Observing that "a State's highest court is the final judicial arbiter of the meaning of state statutes," we said that "[w]hen a state court has made its own definitive determination as to the operating incidence, . . . [w]e give this finding great weight in determining the natural effect of a statute, and if it is consistent with the statute's reasonable interpretation it will be deemed conclusive." Id., at 208.

2 Even in the rare case in which a State's "manner" of making and construing laws might implicate a structural constraint, Congress, not this Court, is likely the proper governmental entity to enforce that constraint. See U. S. CONST., amend. XII; 3 U. S. C. ¤¤1-15; cf. Ohio ex rel. Davis v. Hildebrant, 241 U. S. 565, 569 (1916) (treating as a nonjusticiable political question whether use of a referendum to override a congressional districting plan enacted by the state legislature violates Art. I, ¤4); Luther v. Borden, 7 How. 1, 42 (1849).

3 "[B]ecause the Framers recognized that state power and identity were essential parts of the federal balance, see The Federalist No. 39, the Constitution is solicitous of the prerogatives of the States, even in an otherwise sovereign federal province. The Constitution . . . grants States certain powers over the times, places, and manner of federal elections (subject to congressional revision), Art. I, ¤4, cl. 1 . . . , and allows States to appoint electors for the President, Art. II, ¤1, cl. 2." U. S. Term Limits, Inc. v. Thornton, 514 U. S. 779, 841-842 (1995) (KENNEDY, J., concurring).

Ê

The remainder of petitioners' claims, which are the focus of the CHIEF JUSTICE's concurrence, raise no significant federal questions. I cannot agree that the CHIEF JUSTICE's unusual review of state law in this case, see ante, at 5-8 (GINSBURG, J., dissenting opinion), is justified by reference either to Art. II, ¤1, or to 3 U. S. C. ¤5. Moreover, even were such review proper, the conclusion that the Florida Supreme Court's decision contravenes federal law is untenable.

While conceding that, in most cases, "comity and respect for federalism compel us to defer to the decisions of state courts on issues of state law," the concurrence relies on some combination of Art. II, ¤1, and 3 U. S. C. ¤5 to justify the majority's conclusion that this case is one of the few in which we may lay that fundamental principle aside. Ante, at 2 (Opinion of REHNQUIST, C. J. The concurrence's primary foundation for this conclusion rests on an appeal to plain text: Art. II, ¤1's grant of the power to appoint Presidential electors to the State "Legislature." Ibid. But neither the text of Article II itself nor the only case the concurrence cites that interprets Article II, McPherson v. Blacker, 146 U. S. 1 (1892), leads to the conclusion that Article II grants unlimited power to the legislature, devoid of any state constitutional limitations, to select the manner of appointing electors. See id., at 41 (specifically referring to state constitutional provision in upholding state law regarding selection of electors). Nor, as JUSTICE STEVENS points out, have we interpreted the Federal constitutional provision most analogous to Art. II, ¤1 - Art. I, ¤4 -- in the strained manner put forth in the concurrence. Ante, at 1-2 and n. 1 (dissenting opinion).

The concurrence's treatment of ¤5 as "inform[ing]" its interpretation of Article II, ¤1, cl. 2, ante, at 3 (REHNQUIST, C. J., concurring), is no more convincing. The CHIEF JUSTICE contends that our opinion in Bush v. Palm Beach County Canvassing Bd., ante, p. ____, (per curiam) (Bush I), in which we stated that "a legislative wish to take advantage of [¤5] would counsel against" a construction of Florida law that Congress might deem to be a change in law, id., (slip op. at 6), now means that this Court "must ensure that post-election state court actions do not frustrate the legislative desire to attain the 'safe harbor'provided by ¤5." Ante, at 3. However, ¤5 is part of the rules that govern Congress' recognition of slates of electors. Nowhere in Bush I did we establish that this Court had the authority to enforce ¤5. Nor did we suggest that the permissive "counsel against" could be transformed into the mandatory "must ensure." And nowhere did we intimate, as the concurrence does here, that a state court decision that threatens the safe harbor provision of ¤5 does so in violation of Article II. The concurrence's logic turns the presumption that legislatures would wish to take advantage of ¤ 5's "safe harbor" provision into a mandate that trumps other statutory provisions and overrides the intent that the legislature did express.

But, in any event, the concurrence, having conducted its review, now reaches the wrong conclusion. It says that "the Florida Supreme Court's interpretation of the Florida election laws impermissibly distorted them beyond what a fair reading required, in violation of Article II." Ante, at 4-5 (REHNQUIST, C. J, concurring). But what precisely is the distortion? Apparently, it has three elements. First, the Florida court, in its earlier opinion, changed the election certification date from November 14 to November 26. Second, the Florida court ordered a manual recount of "undercounted" ballots that could not have been fully completed by the December 12 "safe harbor" deadline. Third, the Florida court, in the opinion now under review, failed to give adequate deference to the determinations of canvassing boards and the Secretary.

To characterize the first element as a "distortion," however, requires the concurrence to second-guess the way in which the state court resolved a plain conflict in the language of different statutes. Compare Fla. Stat. ¤102.166 (2001) (foreseeing manual recounts during the protest period) with ¤102.111 (setting what is arguably too short a deadline for manual recounts to be conducted); compare ¤102.112(1) (stating that the Secretary "may" ignore late returns) with ¤102.111(1) (stating that the Secretary "shall" ignore late returns). In any event, that issue no longer has any practical importance and cannot justify the reversal of the different Florida court decision before us now.

To characterize the second element as a "distortion" requires the concurrence to overlook the fact that the inability of the Florida courts to conduct the recount on time is, in significant part, a problem of the Court's own making. The Florida Supreme Court thought that the recount could be completed on time, and, within hours, the Florida Circuit Court was moving in an orderly fashion to meet the deadline. This Court improvidently entered a stay. As a result, we will never know whether the recount could have been completed.

Nor can one characterize the third element as "impermissibl[e] distort[ing]" once one understands that there are two sides to the opinion's argument that the Florida Supreme Court "virtually eliminated the Secretary's discretion." Ante, at 9 (REHNQUIST, C. J, concurring). The Florida statute in question was amended in 1999 to provide that the "grounds for contesting an election" include the "rejection of a number of legal votes sufficient to . . . place in doubt the result of the election." Fla. Stat. ¤¤102.168(3), (3)(c) (2000). And the parties have argued about the proper meaning of the statute's term "legal vote." The Secretary has claimed that a "legal vote" is a vote "properly executed in accordance with the instructions provided to all registered voters." Brief for Respondent Harris et al. 10. On that interpretation, punchcard ballots for which the machines cannot register a vote are not "legal" votes. Id., at 14. The Florida Supreme Court did not accept her definition. But it had a reason. Its reason was that a different provision of Florida election laws (a provision that addresses damaged or defective ballots) says that no vote shall be disregarded "if there is a clear indication of the intent of the voter as determined by the canvassing board" (adding that ballots should not be counted "if it is impossible to determine the elector's choice"). Fla. Stat. ¤101.5614(5) (2000). Given this statutory language, certain roughly analogous judicial precedent, e.g., Darby v. State ex rel. McCollough, 75 So. 411 (Fla. 1917) (per curiam), and somewhat similar determinations by courts throughout the Nation, see cases cited infra, at 9, the Florida Supreme Court concluded that the term "legal vote" means a vote recorded on a ballot that clearly reflects what the voter intended. Gore v. Harris, ___ So. 2d ___, ___ (2000) (slip op., at 19). That conclusion differs from the conclusion of the Secretary. But nothing in Florida law requires the Florida Supreme Court to accept as determinative the Secretary's view on such a matter. Nor can one say that the Court's ultimate determination is so unreasonable as to amount to a constitutionally "impermissible distort[ion]"of Florida law.

The Florida Supreme Court, applying this definition, decided, on the basis of the record, that respondents had shown that the ballots undercounted by the voting machines contained enough "legal votes"to place "the results" of the election "in doubt." Since only a few hundred votes separated the candidates, and since the "undercounted" ballots numbered tens of thousands, it is difficult to see how anyone could find this conclusion unreasonable-however strict the standard used to measure the voter's "clear intent." Nor did this conclusion "strip" canvassing boards of their discretion. The boards retain their traditional discretionary authority during the protest period. And during the contest period, as the court stated, "the Canvassing Board's actions [during the protest period] may constitute evidence that a ballot does or does not qualify as a legal vote." Id., at *13. Whether a local county canvassing board's discretionary judgment during the protest period not to conduct a manual recount will be set aside during a contest period depends upon whether a candidate provides additional evidence that the rejected votes contain enough "legal votes" to place the outcome of the race in doubt. To limit the local canvassing board's discretion in this way is not to eliminate that discretion. At the least, one could reasonably so believe.

The statute goes on to provide the Florida circuit judge with authority to "fashion such orders as he or she deems necessary to ensure that each allegation . . . is investigated, examined, or checked, . . . and to provide any relief appropriate." Fla. Stat. ¤102.168(8) (2000) (emphasis added). The Florida Supreme Court did just that. One might reasonably disagree with the Florida Supreme Court's interpretation of these, or other, words in the statute. But I do not see how one could call its plain language interpretation of a 1999 statutory change so misguided as no longer to qualify as judicial interpretation or as a usurpation of the authority of the State legislature. Indeed, other state courts have interpreted roughly similar state statutes in similar ways. See, e.g., In re Election of U. S. Representative for Second Congressional Dist., 231 Conn. 602, 621, 653 A. 2d 79, 90-91 (1994) ("Whatever the process used to vote and to count votes, differences in technology should not furnish a basis for disregarding the bedrock principle that the purpose of the voting process is to ascertain the intent of the voters"); Brown v. Carr, 130 W. Va. 401, 460, 43 S. E.2d 401, 404-405 (1947) ("[W]hether a ballot shall be counted . . . depends on the intent of the voter . . . . Courts decry any resort to technical rules in reaching a conclusion as to the intent of the voter").

I repeat, where is the "impermissible"distortion?

II

Despite the reminder that this case involves "an election for the President of the United States," ante, at 1 (REHNQUIST, C. J., concurring), no preeminent legal concern, or practical concern related to legal questions, required this Court to hear this case, let alone to issue a stay that stopped Florida's recount process in its tracks. With one exception, petitioners'claims do not ask us to vindicate a constitutional provision designed to protect a basic human right. See, e.g., Brown v. Board of Education, 347 U. S. 483 (1954). Petitioners invoke fundamental fairness, namely, the need for procedural fairness, including finality. But with the one "equal protection" exception, they rely upon law that focuses, not upon that basic need, but upon the constitutional allocation of power. Respondents invoke a competing fundamental consideration -- the need to determine the voter's true intent. But they look to state law, not to federal constitutional law, to protect that interest. Neither side claims electoral fraud, dishonesty, or the like. And the more fundamental equal protection claim might have been left to the state court to resolve if and when it was discovered to have mattered. It could still be resolved through a remand conditioned upon issuance of a uniform standard; it does not require reversing the Florida Supreme Court.

Of course, the selection of the President is of fundamental national importance. But that importance is political, not legal. And this Court should resist the temptation unnecessarily to resolve tangential legal disputes, where doing so threatens to determine the outcome of the election.

The Constitution and federal statutes themselves make clear that restraint is appropriate. They set forth a road map of how to resolve disputes about electors, even after an election as close as this one. That road map foresees resolution of electoral disputes by state courts. See 3 U. S. C. ¤5 (providing that, where a "State shall have provided, by laws enacted prior to [election day], for its final determination of any controversy or contest concerning the appointment of . . . electors . . . by judicial or other methods," the subsequently chosen electors enter a safe harbor free from congressional challenge). But it nowhere provides for involvement by the United States Supreme Court.

To the contrary, the Twelfth Amendment commits to Congress the authority and responsibility to count electoral votes. A federal statute, the Electoral Count Act, enacted after the close 1876 Hayes-Tilden Presidential election, specifies that, after States have tried to resolve disputes (through "judicial" or other means), Congress is the body primarily authorized to resolve remaining disputes. See Electoral Count Act of 1887, 24 Stat. 373, 3 U. S. C. ¤¤5, 6, and 15.

The legislative history of the Act makes clear its intent to commit the power to resolve such disputes to Congress, rather than the courts:
"The two Houses are, by the Constitution, authorized to make the count of electoral votes. They can only count legal votes, and in doing so must determine, from the best evidence to be had, what are legal votes .... The power to determine rests with the two Houses, and there is no other constitutional tribunal." H. Rep. No. 1638, 49th Cong., 1st Sess., 2 (1886) (report submitted by Rep. Caldwell, Select Committee on the Election of President and Vice-President).

The Member of Congress who introduced the Act added:
"The power to judge of the legality of the votes is a necessary consequent of the power to count. The existence of this power is of absolute necessity to the preservation of the Government. The interests of all the States in their relations to each other in the Federal Union demand that the ultimate tribunal to decide upon the election of President should be a constituent body, in which the States in their federal relationships and the people in their sovereign capacity should be represented." 18 Cong. Rec. 30 (1886). "Under the Constitution who else could decide? Who is nearer to the State in determining a question of vital importance to the whole union of States than the constituent body upon whom the Constitution has devolved the duty to count the vote?" Id., at 31.

The Act goes on to set out rules for the congressional determination of disputes about those votes. If, for example, a state submits a single slate of electors, Congress must count those votes unless both Houses agree that the votes "have not been . . . regularly given." 3 U. S. C. ¤ 15. If, as occurred in 1876, one or more states submits two sets of electors, then Congress must determine whether a slate has entered the safe harbor of ¤5, in which case its votes will have "conclusive" effect. Ibid. If, as also occurred in 1876, there is controversy about "which of two or more of such State authorities . . . is the lawful tribunal" authorized to appoint electors, then each House shall determine separately which votes are "supported by the decision of such State so authorized by its law." Ibid. If the two Houses of Congress agree, the votes they have approved will be counted. If they disagree, then "the votes of the electors whose appointment shall have been certified by the executive of the State, under the seal thereof, shall be counted." Ibid.

Given this detailed, comprehensive scheme for counting electoral votes, there is no reason to believe that federal law either foresees or requires resolution of such a political issue by this Court. Nor, for that matter, is there any reason to that think the Constitution's Framers would have reached a different conclusion. Madison, at least, believed that allowing the judiciary to choose the presidential electors "was out of the question." Madison, July 25, 1787 (reprinted in 5 Elliot's Debates on the Federal Constitution 363 (2d ed. 1876)).

The decision by both the Constitution's Framers and the 1886 Congress to minimize this Court's role in resolving close federal presidential elections is as wise as it is clear. 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.

Moreover, Congress was fully aware of the danger that would arise should it ask judges, unarmed with appropriate legal standards, to resolve a hotly contested Presidential election contest. Just after the 1876 Presidential election, Florida, South Carolina, and Louisiana each sent two slates of electors to Washington. Without these States, Tilden, the Democrat, had 184 electoral votes, one short of the number required to win the Presidency. With those States, Hayes, his Republican opponent, would have had 185. In order to choose between the two slates of electors, Congress decided to appoint an electoral commission composed of five Senators, five Representatives, and five Supreme Court Justices. Initially the Commission was to be evenly divided between Republicans and Democrats, with Justice David Davis, an Independent, to possess the decisive vote. However, when at the last minute the Illinois Legislature elected Justice Davis to the United States Senate, the final position on the Commission was filled by Supreme Court Justice Joseph P. Bradley.

The Commission divided along partisan lines, and the responsibility to cast the deciding vote fell to Justice Bradley. He decided to accept the votes by the Republican electors, and thereby awarded the Presidency to Hayes.

Justice Bradley immediately became the subject of vociferous attacks. Bradley was accused of accepting bribes, of being captured by railroad interests, and of an eleventh-hour change in position after a night in which his house "was surrounded by the carriages" of Republican partisans and railroad officials. C. Woodward, Reunion and Reaction 159-160 (1966). Many years later, Professor Bickel concluded that Bradley was honest and impartial. He thought that "'the great question'for Bradley was, in fact, whether Congress was entitled to go behind election returns or had to accept them as certified by state authorities,"an "issue of principle." The Least Dangerous Branch 185 (1962). Nonetheless, Bickel points out, the legal question upon which Justice Bradley's decision turned was not very important in the contemporaneous political context. He says that "in the circumstances the issue of principle was trivial, it was overwhelmed by all that hung in the balance, and it should not have been decisive."Ibid.

For present purposes, the relevance of this history lies in the fact that the participation in the work of the electoral commission by five Justices, including Justice Bradley, did not lend that process legitimacy. Nor did it assure the public that the process had worked fairly, guided by the law. Rather, it simply embroiled Members of the Court in partisan conflict, thereby undermining respect for the judicial process. And the Congress that later enacted the Electoral Count Act knew it.

This history may help to explain why I think it not only legally wrong, but also most unfortunate, for the Court simply to have terminated the Florida recount. Those who caution judicial restraint in resolving political disputes have described the quintessential case for that restraint as a case marked, among other things, by the "strangeness of the issue," its "intractability to principled resolution," its "sheer momentousness, . . . which tends to unbalance judicial judgment," and "the inner vulnerability, the self-doubt of an institution which is electorally irresponsible and has no earth to draw strength from." Bickel, supra, at 184. Those characteristics mark this case.

At the same time, as I have said, the Court is not acting to vindicate a fundamental constitutional principle, such as the need to protect a basic human liberty. No other strong reason to act is present. Congressional statutes tend to obviate the need. And, above all, in this highly politicized matter, the appearance of a split decision runs the risk of undermining the public's confidence in the Court itself. That confidence is a public treasure. It has been built slowly over many years, some of which were marked by a Civil War and the tragedy of segregation. It is a vitally necessary ingredient of any successful effort to protect basic liberty and, indeed, the rule of law itself. We run no risk of returning to the days when a President (responding to this Court's efforts to protect the Cherokee Indians) might have said, "John Marshall has made his decision; now let him enforce it!" Loth, Chief Justice John Marshall and The Growth of the American Republic 365 (1948). But we do risk a self-inflicted wound -- a wound that may harm not just the Court, but the Nation.

I fear that in order to bring this agonizingly long election process to a definitive conclusion, we have not adequately attended to that necessary "check upon our own exercise of power," "our own sense of self-restraint."