Opinion

JUSTICE O'CONNOR, with whom THE CHIEF JUSTICE and JUSTICE REHNQUIST join, concurring in the judgment.

Today the Court holds that claims of political gerrymandering lodged by members of one of the political parties that make up our two-party system are justiciable under the Equal Protection Clause of the Fourteenth Amendment. Nothing in our precedents compels us to take this step, and there is every reason not to do so. I would hold that the partisan gerrymandering claims of major political parties raise a nonjusticiable political question that the judiciary should leave to the legislative branch, as the Framers of the Constitution unquestionably intended. Accordingly, I would reverse the District Court's judgment on the grounds that appellees' claim is nonjusticiable.

There can be little doubt that the emergence of a strong and stable two-party system in this country has contributed enormously to sound and effective government. The preservation and health of our political institutions, state and federal, depends to no small extent on the continued vitality of our two-party system, which permits both stability and measured change. The opportunity to control the drawing of electoral boundaries through the legislative process of apportionment is a critical and traditional part of politics in the United States, and one that plays no small role in fostering active participation in the political parties at every level. Thus, the legislative business of apportionment is fundamentally a political affair, and challenges to the manner in which an apportionment has been carried out -by the very parties that are responsible for this process -present a political question in the truest sense of the term.

To turn these matters over to the federal judiciary is to inject the courts into the most heated partisan issues. It is predictable that the courts will respond by moving away from the nebulous standard a plurality of the Court fashions today and toward some form of rough proportional representation for all political groups. The consequences of this shift will be as immense as they are unfortunate. I do not believe, and the Court offers not a shred of evidence to suggest, that the Framers of the Constitution intended the judicial power to encompass the making of such fundamental choices about how this Nation is to be governed. Nor do I believe that the proportional representation towards which the Court's expansion of equal protection doctrine will lead is consistent with our history, our traditions, or our political institutions.

The Court pays little heed to these considerations, which should inform any sensible jurisprudence of Article III and of the Equal Protection Clause. The Court's reflexive application of precedent ignores the maxim that,

[p]articularly in dealing with claims under broad provisions of the Constitution, which derive content by an interpretative process of inclusion and exclusion, it is imperative that generalizations, based on and qualified by the concrete situations that gave rise to them, must not be applied out of context in disregard of variant controlling facts.

Gomillion v. Lightfoot, 364 U. S. 339, 364 U. S. 343 -344 (1960). In cases such as this one, which may profoundly affect the governance of this Nation, it is not enough to cite precedent: we should examine it for possible limits, and, if they are lacking, for possible flaws.

I

Appellees are Indiana Democrats who claim that Indiana's 1981 state apportionment discriminates against Democrats on a statewide basis by diluting their votes, thereby depriving them of "their proportionate share of political influence." Baker v. Carr, 369 U. S. 186, 369 U. S. 299 (1962) (Frankfurter, J., dissenting). The Court, relying principally on Baker v. Carr, supra, Reynolds v. Sims, 377 U. S. 533 (1964), Gaffney v. Cummings, 412 U. S. 735 (1973), and the line of racial gerrymandering cases including Rogers v. Lodge, 458 U. S. 613 (1982), and White v. Regester, 412 U. S. 755 (1973), holds that appellees"'purely political equal protection claim," ante at 119, does not present a political question, and is therefore justiciable. Specifically, the Court holds that the fact that a vote dilution claim "is submitted by a political group, rather than a racial group, does not distinguish it in terms of justiciability." Ante at 478 U. S. 125.

A plurality of the Court recognizes, however, that

[i]nviting attack on minor departures from some supposed norm would too much embroil the judiciary in second-guessing what has consistently been referred to as a political task for the legislature, a task that should not be monitored too closely unless the express or tacit goal is to effect its removal from legislative halls.

Ante at 478 U. S. 133. Accordingly, although the plurality's analysis is generally modeled on the racial gerrymandering cases, the plurality would require a somewhat different threshold showing that the apportionment has discriminatory effects:

unconstitutional discrimination occurs only when the electoral system is arranged in a manner that will consistently degrade a voter's or a group of voters' influence on the political process as a whole.

Ante at 478 U. S. 132.

The step taken today is a momentous one which, if followed in the future, can only lead to political instability and judicial malaise. If members of the major political parties are protected by the Equal Protection Clause from dilution of their voting strength, then members of every identifiable group that possesses distinctive interests and tends to vote on the basis of those interests should be able to bring similar claims. Federal courts will have no alternative but to attempt to recreate the complex process of legislative apportionment in the context of adversary litigation in order to reconcile the competing claims of political, religious, ethnic, racial, occupational, and socioeconomic groups. Even if there were some way of limiting such claims to organized political parties, the fact remains that the losing party or the losing group of legislators in every reapportionment will now be invited to fight the battle anew in federal court. Apportionment is so important to legislators and political parties that the burden of proof the plurality places on political gerrymandering plaintiffs is unlikely to deter the routine lodging of such complaints. Notwithstanding the plurality's threshold requirement of discriminatory effects, the Court's holding that political gerrymandering claims are justiciable has opened the door to pervasive and unwarranted judicial superintendence of the legislative task of apportionment. There is simply no clear stopping point to prevent the gradual evolution of a requirement of roughly proportional representation for every cohesive political group.

In my view, this enterprise is flawed from its inception. The Equal Protection Clause does not supply judicially manageable standards for resolving purely political gerrymandering claims, and no group right to an equal share of political power was ever intended by the Framers of the Fourteenth Amendment. The Court rests its case on precedent, but the cases on which the Court relies do not require that we take this next and most far-reaching step into the "political thicket." Colegrove v. Green, 328 U. S. 549, 328 U. S. 556 (1946) (opinion of Frankfurter, J.).

Baker v. Carr reaffirmed that a lawsuit will be held to involve a political question where there is "a lack of judicially discoverable and manageable standards for resolving it," or where "the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion" is apparent. 369 U.S. at 369 U. S. 217. The Court first found a workable constitutional standard for applying the Equal Protection Clause to state legislative districting in Reynolds v. Sims, supra. But, until today, the Court has not extended the principles of Baker v. Carr and Reynolds v. Sims to test a legislative districting plan on grounds of partisan political gerrymandering. Indeed, one year after Reynolds v. Sims, the Court was unanimous in summarily affirming a judgment determining that a political gerrymandering challenge was nonjusticiable; as Justice Harlan pointed out, the Court's action constituted a rejection of

contentions that... partisan 'gerrymandering' may be subject to federal constitutional attack under the Fourteenth Amendment.

WMCA, Inc. v. Lomenzo, 382 U. S. 4, 382 U. S. 6 (1965) (concurring opinion).

The question raised again today, 21 years later, is whether a court can apply the familiar "[j]udicial standards under the Equal Protection Clause," Baker v. Carr, 369 U.S. at 369 U. S. 226, without being forced to make a nonjudicial policy determination or to resort to a standard that is not judicially manageable. In order to answer that question, it is necessary to interpret the Equal Protection Clause. As Justice Harlan pointed out in his dissent in Baker v. Carr,

[t]he suggestion... that courts lack standards by which to decide such cases as this is relevant not only to the question of 'justiciability,' but also, and perhaps more fundamentally, to the determination whether any cognizable constitutional claim has been asserted in this case.

Id. at 369 U. S. 337. Baker v. Carr itself involved just such an initial interpretation of the Equal Protection Clause: the Court in effect ruled that an arbitrary and capricious discrimination against individual voters with respect to the weight of their votes would state a cognizable claim under the Equal Protection Clause. See id. at 369 U. S. 226 ; id. at 369 U. S. 338 -339 (Harlan, J., dissenting). That threshold determination about the reach and meaning of the Equal Protection Clause was the basis for the Court's holding that the complaint of the Tennessee voters was justiciable. Even this "arbitrary and capricious" standard threatened to prove unmanageable, but the difficulty was pretermitted when a relatively simple and judicially manageable requirement of population equality among districts was adopted the following Term in Reynolds v. Sims. See Bickel, The Supreme Court and Reapportionment, in Reapportionment in the 1970's, pp. 57, 64 (N. Polsby ed.1971).

Baker v. Carr does not require that we hold that the right asserted in this case is similarly within the intendment of the Equal Protection Clause and determinable under the standards developed to enforce that Clause. The right asserted in Baker v. Carr was an individual right to a vote whose weight was not arbitrarily subjected to "debasement," 369 U.S. at 369 U. S. 194. The rights asserted in this case are group rights to an equal share of political power and representation, and the "arbitrary and capricious" standard discussed in Baker v. Carr cannot serve as the basis for recognizing such rights. Indeed, the Court today does not rely on such a standard.

Instead, the Court justifies the extension of vote dilution claims to mainstream political groups with the pronouncement that " Reynolds surely indicates the justiciability of claims going to the adequacy of representation in state legislatures." Ante at 478 U. S. 124. But Reynolds makes plain that the one person, one vote principle safeguards the individual's right to vote, not the interests of political groups:

To the extent that a citizen's right to vote is debased, he is that much less a citizen. The fact that an individual lives here or there is not a legitimate reason for overweighting or diluting the efficacy of his vote.

377 U.S. at 377 U. S. 567. For that reason,

an individual's right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the State.

Id.. at 377 U. S. 568. Thus, the right guaranteed by the Equal Protection Clause as interpreted in Reynolds is "the right of each voter to have his vote weighted equally with those of all other citizens.'" Mobile v. Bolden, 446 U. S. 55, 446 U. S. 78 (1980) (plurality opinion).

In the case of mainstream political groups, the Court has not accepted the argument that an "asserted entitlement to group representation," Bolden, 446 U.S. at 446 U. S. 77, can be traced to the one person, one vote principle:

It is, of course, true that the right of a person to vote on an equal basis with other voters draws much of its significance from the political associations that its exercise reflects, but it is an altogether different matter to conclude that political groups themselves have an independent constitutional claim to representation. And the Court's decisions hold squarely that they do not.

Id. at 446 U. S. 78 -79 (citing United Jewish Organizations of Williamsburgh, Inc. v. Carey, 430 U. S. 144 (1977); White v. Regester, 412 U. S. 755 (1973); and Whitcomb v. Chavis, 403 U. S. 124 (1971)). Where representation is apportioned substantially on a population basis, it is "obvious that nobody's vote has been diluted' in the sense in which that word was used in the Reynolds case." Bolden, supra, at 446 U. S. 78. Thus, the individual's right to vote does not imply that political groups have a right to be free from discriminatory impairment of their group voting strength. Treating the vote dilution claims of political groups as cognizable would effectively collapse the

fundamental distinction between state action that inhibits an individual's right to vote and state action that affects the political strength of various groups that compete for leadership in a democratically governed community.

Bolden, supra, at 446 U. S. 83 (STEVENS, J., concurring in judgment).

Nor do this Court's racial gerrymandering cases require the recognition of any such group right outside the context of racial discrimination. As Justice Frankfurter observed:

The cases involving Negro disfranchisement are no exception to the principle of avoiding federal judicial intervention into matters of state government in the absence of an explicit and clear constitutional imperative. For here the controlling command of Supreme Law is plain and unequivocal. An end of discrimination against the Negro was the compelling motive of the Civil War Amendments. The Fifteenth expresses this in terms, and it is no less true of the Equal Protection Clause of the Fourteenth. Slaughter-House Cases, 16 Wall. 36, 83 U. S. 67 -72; Strauder v. West Virginia, 100 U. S. 303, 100 U. S. 306 -307; Nixon v. Herndon, 273 U. S. 536, 273 U. S. 541.

Baker v. Carr, 369 U.S. at 369 U. S. 285 -286 (dissenting opinion).

In my view, where a racial minority group is characterized by "the traditional indicia of suspectness" and is vulnerable to exclusion from the political process, San Antonio Independent School District v. Rodriguez, 411 U. S. 1, 411 U. S. 28 (1973); see also Johnson v. Robison, 415 U. S. 361, 415 U. S. 375, n. 14 (1974), individual voters who belong to that group enjoy some measure of protection against intentional dilution of their group voting strength by means of racial gerrymandering. As a matter of past history and present reality, there is a direct and immediate relationship between the racial minority's group voting strength in a particular community and the individual rights of its members to vote and to participate in the political process. In these circumstances, the stronger nexus between individual rights and group interests, and the greater warrant the Equal Protection Clause gives the federal courts to intervene for protection against racial discrimination, suffice to render racial gerrymandering claims justiciable. Even so, the individual's right is infringed only if the racial minority group can prove that it has "essentially been shut out of the political process." Ante at 478 U. S. 139.

Clearly, members of the Democratic and Republican Parties cannot claim that they are a discrete and insular group vulnerable to exclusion from the political process by some dominant group: these political parties are the dominant groups, and the Court has offered no reason to believe that they are incapable of fending for themselves through the political process. Indeed, there is good reason to think that political gerrymandering is a self-limiting enterprise. See B. Cain, The Reapportionment Puzzle 151-159 (1984). In order to gerrymander, the legislative majority must weaken some of its safe seats, thus exposing its own incumbents to greater risks of defeat -risks they may refuse to accept past a certain point. Id. at 154-155. Similarly, an overambitious gerrymander can lead to disaster for the legislative majority: because it has created more seats in which it hopes to win relatively narrow victories, the same swing in overall voting strength will tend to cost the legislative majority more and more seats as the gerrymander becomes more ambitious. Id. at 152. More generally, each major party presumably has ample weapons at its disposal to conduct the partisan struggle that often leads to a partisan apportionment, but also often leads to a bipartisan one. There is no proof before us that political gerrymandering is an evil that cannot be checked or cured by the people or by the parties themselves. Absent such proof, I see no basis for concluding that there is a need, let alone a constitutional basis, for judicial intervention.

The plurality agrees that it would be unwise to "embroil the judiciary in second-guessing what has consistently been referred to as a political task for the legislature." Ante at 478 U. S. 133. Moreover, the plurality is willing to presume that elected candidates will not ignore the interests of voters for the losing candidate, and it correctly observes that "the power to influence the political process is not limited to winning elections." Ante at 478 U. S. 132. But these propositions support my position -that the costs of judicial intervention will be severe, and that political gerrymandering simply does not cause intolerable harm to the ability of major political groups to advance their interests.

Moreover, the new group right created by today's decision is particularly unjustifiable in the context of the claim here, which is founded on a supposed diminution of the statewide voting influence of a political group. None of the elections for the Indiana Legislature are statewide. Voters in each district elect their representatives from that district. To treat the loss of candidates nominated by the party of a voter's choice as a harm to the individual voter, when that voter cannot vote for such candidates and is not represented by them in any direct sense, clearly exceeds the limits of the Equal Protection Clause. On the Court's reasoning, members of a political party in one State should be able to challenge a congressional districting plan adopted in any other State, on the grounds that their party is unfairly represented in that State's congressional delegation, thus injuring them as members of the national party.

The Court's reliance on Gaffney v. Cummings, 412 U. S. 735 (1973), is insufficient to overcome these objections to a general group right to equal political representation. Although Gaffney treated a political gerrymandering claim as justiciable, the opinion's observation that "districting inevitably has and is intended to have substantial political consequences," id. at 412 U. S. 753, and its reluctance to undertake "the impossible task of extirpating politics from what are the essentially political processes of the sovereign States," id. at 412 U. S. 754, would equally support a holding that whatever harms political gerrymandering may sometimes occasion should be tolerated as inextricably associated with the legislative business of redistricting. In addition, since Gaffney rejected the challenge to bipartisan gerrymandering out of hand, the Court simply did not confront the difficulties in framing a manageable standard for adjudicating such claims. Accordingly, Gaffney should not bar a full consideration of those difficulties here.

Furthermore, the Court fails to explain why a bipartisan gerrymander -which is what was approved in Gaffney -affects individuals any differently than a partisan gerrymander, which the Court makes vulnerable to constitutional challenge today. In Gaffney, Connecticut, as part of a bipartisan effort, had drawn up a plan intended to "provide a rough sort of proportional representation," id. at 412 U. S. 754, for the two major political parties. The Court declined to invalidate this plan, which undertook "not to minimize or eliminate the political strength of any group or party, but to recognize it," ibid., and suggested that

judicial interest should be at its lowest ebb when a State purports fairly to allocate political power to the parties in accordance with their voting strength and, within quite tolerable limits, succeeds in doing so.

Ibid. (citations omitted).

A bipartisan gerrymander employs the same technique, and has the same effect on individual voters, as does a partisan gerrymander. In each instance, groups of individuals are assigned to districts with an eye towards promoting the ends of a political party and its incumbent legislators. Some groups within each party will lose any chance to elect a representative who belongs to their party, because they have been assigned to a district in which the opposing party holds an overwhelming advantage. Independent voters may lose any chance to influence the outcome of elections in their district, if one party has a sufficiently strong majority. As the plurality acknowledges, the scheme upheld in Gaffney tended to "deny safe district minorities any realistic chance to elect their own representatives." Ante at 478 U. S. 131. If this bipartisan arrangement between two groups of self-interested legislators is constitutionally permissible, as I believe and as the Court held in Gaffney, then -in terms of the rights of individuals -it should be equally permissible for a legislative majority to employ the same means to pursue its own interests over the opposition of the other party.

The Court's determination to treat the claims of mainstream political parties as justiciable thus emerges as precisely the sort of "initial policy determination of a kind clearly for nonjudicial discretion" that Baker v. Carr recognized as characteristic of political questions. 369 U.S. at 369 U. S. 217. The Court has in effect decided that it is constitutionally acceptable for both parties to "waste" the votes of individuals through a bipartisan gerrymander, so long as the parties themselves are not deprived of their group voting strength to an extent that will exceed the plurality's threshold requirement. This choice confers greater rights on powerful political groups than on individuals; that cannot be the meaning of the Equal Protection Clause.

II

The standard the plurality proposes exemplifies the intractable difficulties in deriving a judicially manageable standard from the Equal Protection Clause for adjudicating political gerrymandering claims. The plurality rejects any standard that would require drawing

district lines to come as near as possible to allocating seats to the contending parties in proportion to what their anticipated statewide vote will be,

ante at 478 U. S. 130, and states that

unconstitutional discrimination occurs only when the electoral system is arranged in a manner that will consistently degrade a voter's or a group of voters' influence on the political process as a whole.

Ante at 478 U. S. 132. In my view, this standard will, over time, either prove unmanageable and arbitrary or else evolve towards some loose form of proportionality. Cf. Shapiro, Gerrymandering, Unfairness, and the Supreme Court, 33 UCLA L.Rev. 227, 252-256 (1985). Either outcome would be calamitous for the federal courts, for the States, and for our two-party system.

Vote dilution analysis is far less manageable when extended to major political parties than if confined to racial minority groups. First, an increase in the number of competing claims to equal group representation will make judicial review of apportionment vastly more complex. Designing an apportionment plan that does not impair or degrade the voting strength of several groups is more difficult than designing a plan that does not have such an effect on one group for the simple reason that, as the number of criteria the plan must meet increases, the number of solutions that will satisfy those criteria will decrease. Even where it is not impossible to reconcile the competing claims of political, racial, and other groups, the predictable result will be greater judicial intrusion into the apportionment process.

Second, while membership in a racial group is an immutable characteristic, voters can -and often do -move from one party to the other or support candidates from both parties. Consequently, the difficulty of measuring voting strength is heightened in the case of a major political party. It is difficult enough to measure "a voter's or a group of voters' influence on the political process as a whole," ante at 478 U. S. 132, when the group is a racial minority in a particular district or community. When the group is a major political party, the difficulty is greater, and the constitutional basis for intervening far more tenuous.

Moreover, any such intervention is likely to move in the direction of proportional representation for political parties. This is clear by analogy to the problem that arises in racial gerrymandering cases:

in order to decide whether an electoral system has made it harder for minority voters to elect the candidates they prefer, a court must have an idea in mind of how hard it 'should' be for minority voters to elect their preferred candidates under an acceptable system.

Thornburg v. Gingles, ante at 478 U. S. 88 (O'CONNOR, J., concurring in judgment). Any such norm must make some reference, even if only a loose one, to the relation between the racial minority group's share of the electorate and its share of the elected representatives. In order to implement the plurality's standard, it will thus be necessary for courts to adopt an analogous norm, in order to assess whether the voting strength of a political party has been "degraded" by an apportionment, either on a statewide basis or in particular districts. Absent any such norm, the inquiry the plurality proposes would be so standardless as to make the adjudication of political gerrymandering claims impossible.

Implicit in the plurality's opinion today is at least some use of simple proportionality as the standard for measuring the normal representational entitlements of a political party. That is why the plurality can say that "a history (actual or projected) of disproportionate results," together with proof of "the denial of fair representation" and of "lack of political power," will constitute an equal protection violation. Ante at 478 U. S. 139. To be sure, the plurality has qualified its use of a standard of proportional representation in a variety of ways so as to avoid a requirement of proportional representation. The question is whether these qualifications are likely to be enduring in the face of the tremendous political pressures that courts will confront when called on to decide political gerrymandering claims. Because the most easily measured indicia of political power relate solely to winning and losing elections, there is a grave risk that the plurality's various attempts to qualify and condition the group right the Court has created will gradually pale in importance. What is likely to remain is a loose form of proportionality, under which some deviations from proportionality are permissible, but any significant, persistent deviations from proportionality are suspect. Courts will be forced to look for some form of "undue" disproportionality with respect to electoral success if political gerrymandering claims are justiciable, because otherwise they will find their decisions turning on imponderables such as whether the legislators of one party have fairly represented the voters of the other.

Of course, in one sense, a requirement of proportional representation, whether loose or absolute, is judicially manageable. If this Court were to declare that the Equal Protection Clause required proportional representation within certain fixed tolerances, I have no doubt that district courts would be able to apply this edict. The flaw in such a pronouncement, however, would be the use of the Equal Protection Clause as the vehicle for making a fundamental policy choice that is contrary to the intent of its Framers and to the traditions of this Republic. The political question doctrine, as articulated in Baker v. Carr, rightly requires that we refrain from making such policy choices in order to evade what would otherwise be a lack of judicially manageable standards. See 369 U.S. at 369 U. S. 217.

Unfortunately, a drift towards proportional representation is apparent even in the plurality opinion. Although at times the plurality seems to require that the political party be "essentially... shut out of the political process" before a constitutional violation will be found, ante at 478 U. S. 139, the plurality's explanation of the deficiencies in the District Court's approach focuses not on access to the political process as a whole, but entirely on statewide electoral success. Thus, the critical inquiry appears to be into whether the complaining political party could be expected to regain control of the state legislature in the next few elections if backed by a majority of voters. Ante at 478 U. S. 135 -136. As an aid in this inquiry, courts must apparently also ask "by what percentage the statewide... vote" for the complaining political party would have to increase to control the legislature or one of its Houses. Ibid.

Under the plurality's approach, where it is shown that, under a challenged apportionment plan, one party will consistently fail to gain control of the legislature even if it wins a majority of the votes, a court would be justified in finding the "threshold showing" met, at which point "the legislation would be examined for valid underpinnings." Ante at 478 U. S. 141. It may fairly be doubted that this last step is anything more than a formality, except perhaps in the case of bipartisan gerrymanders that have proved unexpectedly favorable to one party. Consequently, although the plurality criticizes JUSTICE POWELL for effectively concluding that "disproportionate election results alone are a sufficient effect to support a finding of a constitutional violation," ante at 478 U. S. 142, the plurality itself arrives at the conclusion that foreseeable, disproportionate long-term election results suffice to prove a constitutional violation.

Thus, the plurality opinion ultimately rests on a political preference for proportionality -not an outright claim that proportional results are required, but a conviction that, the greater the departure from proportionality, the more suspect an apportionment plan becomes. This preference for proportionality is in serious tension with essential features of state legislative elections. Districting itself represents a middle ground between winner-take-all statewide elections and proportional representation for political parties. If there is a constitutional preference for proportionality, the legitimacy of districting itself is called into question: the voting strength of less evenly distributed groups will invariably be diminished by districting as compared to at-large proportional systems for electing representatives. Moreover, one implication of the districting system is that voters cast votes for candidates in their districts, not for a statewide slate of legislative candidates put forward by the parties. Consequently, efforts to determine party voting strength presuppose a norm that does not exist -statewide elections for representatives along party lines.

The plurality's theory is also internally inconsistent. The plurality recognizes that, given a normal dispersion of party strength and winner-take-all, district-based elections, it is likely that even a narrow statewide preference for one party will give that party a disproportionately large majority in the legislature. Ante at 478 U. S. 130. The plurality is prepared to tolerate this effect because not to do so would spell the end of district-based elections, or require reverse gerrymandering to ensure greater proportionality for the minority party. But this means that the plurality would extend greater protection to a party that can command a majority of the statewide vote than to a party that cannot: the explanation, once again, is that the plurality has made a political judgment -in this instance, that district-based elections must be taken as a given.

Because a statewide majority for a party's candidates will frequently result only if the "winning" party attracts independent voters and voters from the other party, under the plurality's approach, a great deal will turn on whether the support of these voters is included as part of the party's voting strength. The plurality would reserve this question, but, however it is ultimately answered, anomalies will result. To measure a party's voting strength by including voters who only occasionally vote for that party's candidates is arbitrary; to ignore the role these voters play will be to further discriminate against parties that do not command a permanent majority of the electorate in a given State.

I would avoid the difficulties generated by the plurality's efforts to confine the effects of a generalized group right to equal representation by not recognizing such a right in the first instance. To allow district courts to strike down apportionment plans on the basis of their prognostications as to the outcome of future elections or future apportionments invites "findings" on matters as to which neither judges nor anyone else can have any confidence. Once it is conceded that "a group's electoral power is not unconstitutionally diminished by the simple fact of an apportionment scheme that makes winning elections more difficult," ante at 478 U. S. 132, the virtual impossibility of reliably predicting how difficult it will be to win an election in 2, or 4, or 10 years should, in my view, weigh in favor of holding such challenges nonjusticiable. Racial gerrymandering should remain justiciable, for the harms it engenders run counter to the central thrust of the Fourteenth Amendment. But no such justification can be given for judicial intervention on behalf of mainstream political parties, and the risks such intervention poses to our political institutions are unacceptable. "Political affiliation is the keystone of the political trade. Race, ideally, is not." United Jewish Organizations of Williamsburgh, Inc. v. Carey, 430 U.S. at 430 U. S. 171, n. 1 (BRENNAN, J., concurring).