Bates v. Jones, 131 F.3d 843 (9th Cir. 1997)

OVERVIEW

A former state legislator and several of his constituents filed this action, contending the lifetime term limits in California's Proposition 140 violate their federal constitutional rights. After a trial, the district court agreed and enjoined the Proposition's enforcement. The district court stayed its injunction pending appeal.

A divided three-judge panel of this court affirmed the district court. A majority of the active judges of the full court then voted to rehear the case en banc and, to accommodate the parties' interests, we agreed to rehear the case on an expedited basis. We have done so and we now reverse the district court.

FACTS

The facts are set forth in detail in the panel's opinion, see Jones v. Bates, 127 F.3d 839 (9th Cir. 1997). We summarize them briefly.

In 1990, California voters approved Proposition 140, an initiative which imposed specific lifetime term limits for state legislators and certain state officers. The Proposition limited state senators to two terms, state assembly members to three terms, and the state governor to two terms. Cal. Const. art. IV, § 2(a); art. V, § 2. The Proposition also limited to two terms the Lieutenant Governor, Attorney General, Controller, Secretary of State, Treasurer, Superintendent of Public Instruction, and the members of the Board of Equalization. Id. at art. V, § 11; art. IX, § 2; art. XII, § 17. The Proposition declared that the lack of term limits created "unfair incumbent advantages" which "discourage qualified candidates from seeking public office and create a class of career politicians, instead of the citizen representatives envisioned by the Founding Fathers." Id. at art. IV, § 1.5. The Proposition stated the term limits were necessary " [t]o restore a free and democratic system of fair elections, and to encourage qualified candidates to seek public office...." Id.

In 1991, the state legislature and several individual legislators and constituents challenged before the California Supreme Court the constitutionality of Proposition 140's term limits. On a petition for a writ of mandate, the California Supreme Court concluded that Proposition 140's lifetime term limits did not violate the plaintiffs' federal constitutional rights. See Legislature v. Eu, 54 Cal. 3d 492, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991), cert. denied, 503 U.S. 919, 112 S. Ct. 1292, 117 L. Ed. 2d 516 (1992).

Thereafter, in 1995, Tom Bates, a former member of the California Assembly, and a group of his constituents filed the present action, also alleging the lifetime term limits of Proposition 140 are unconstitutional. The district court agreed. See Bates v. Jones, 958 F. Supp. 1446 (N.D. Cal. 1997). The district court determined Proposition 140 imposed a severe burden on the plaintiffs' first and fourteenth amendment rights and was not narrowly tailored to advance a compelling state interest. The district court enjoined the enforcement of Proposition 140 but stayed its injunction pending appeal.

A panel of this court, with Judge Sneed dissenting, affirmed the judgment of the district court on other grounds and did not reach the issue whether the term limits are constitutional. Bates, 127 F.3d at 844. This en banc review followed.

DISCUSSION

The State presents a strong argument that res judicata bars the plaintiffs from bringing the present action because they are bound by the decision of the California Supreme Court in Eu. We conclude, however, that California would apply its public interest exception to the res judicata doctrine and, thus, would reexamine the merits of the constitutional issue.

California recognizes an exception to the doctrine of res judicata when "the public interest requires that relitigation not be foreclosed." Kopp v. Fair Political Practices Comm'n, 11 Cal. 4th 607, 47 Cal. Rptr. 2d 108, 115, 905 P.2d 1248, 1256 (1995) (quotations and citations omitted). When the issue previously litigated involves an issue of public importance and there are unusual circumstances favoring reexamination of the issue, California does not apply preclusive effect to the prior determination. See id.; City of Sacramento v. State, 50 Cal. 3d 51, 266 Cal. Rptr. 139, 144, 785 P.2d 522, 528-29 (1990); Arcadia Unified Sch. Dist. v. State Dep't of Educ., 2 Cal. 4th 251, 5 Cal. Rptr. 2d 545, 546, 825 P.2d 438, 440-42 (1992).

The current case justifies application of the public interest exception. In Eu, the California Supreme Court decided to exercise its original jurisdiction on a petition for a writ of mandate, because of the significance and importance of the legal issues raised by the challenge to Proposition 140. As a result, the usual avenues of appellate review were not utilized and the California Supreme Court did not have the benefit of a lower court record. Further, when deciding Eu, there was a paucity of case law addressing the validity of term limits. Since Eu, the United States Supreme Court has decided two significant cases, shedding light on that issue, U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 115 S. Ct. 1842, 131 L. Ed. 2d 881 (1995), and Burdick v. Takushi, 504 U.S. 428, 112 S. Ct. 2059, 119 L. Ed. 2d 245 (1992). We conclude that, in the unique circumstances of this case, the public interest exception applies. We therefore consider the merits of the case.

The three-judge panel did not resolve whether Proposition 140 violates the plaintiffs' first and fourteenth amendment rights. Instead, the panel determined Proposition 140 was invalid because the Proposition and the ballot materials did not provide California voters with sufficient notice that the Proposition imposed lifetime rather than consecutive term limits. Bates, 127 F.3d at 844. We disagree, and, consistent with the California Supreme Court, we hold that the relevant ballot materials and the surrounding context provided sufficient notice making it clear that Proposition 140 required lifetime bans.

The portion of the Proposition affecting legislators states: "No Senator may serve more than 2 terms" and "No member of the Assembly may serve more than 3 terms." Nowhere in the Proposition does it state that these bans are less than absolute. As Judge Sneed pointed out in his dissent from the three-judge panel decision, the twenty-second amendment to the Constitution uses similar language: " [n]o person shall be elected to the office of the President more than twice...." There certainly is no confusion that this language imposes a lifetime ban on the office of the President--even though the amendment does not specifically use the term "lifetime."

The surrounding circumstances also clearly indicate the voters had sufficient notice that Proposition 140 imposed lifetime bans. The opposition materials to the Proposition, which were circulated to California voters, clearly state that elected state legislators will be "banned for life" and use "lifetime ban" or similar terminology no less than eleven times. Moreover, when Proposition 140 was submitted to the voters in 1990, there were two competing initiatives on the ballot imposing term limits. In contrast to Proposition 140's lifetime ban, Proposition 131 proposed consecutive term limits. The two propositions received extensive media attention, which was heightened after the California Supreme Court issued a decision five days before the election addressing which of two propositions would govern in the event both were approved. See Taxpayers to Limit Campaign Spending v. Fair Political Practices Comm., 51 Cal. 3d 744, 274 Cal. Rptr. 787, 799 P.2d 1220 (1990) (specifically addressing Propositions 68 and 73).

Assuming, without deciding, that a federal court may determine whether a state has given adequate notice to its voters in connection with a statewide initiative ballot measure dealing with term limits on state officeholders, we hold that California's notice with regard to Proposition 140 was sufficient.

C. Constitutionality of Proposition 140's Lifetime Term Limits

In Burdick, the Supreme Court set forth the analysis we must apply to determine the constitutionality of Proposition 140. We

must weigh "the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate" against "the precise interests put forward by the State as justifications for the burden imposed by its rule," taking into consideration "the extent to which those interests make it necessary to burden the plaintiff's rights."

Burdick, 504 U.S. at 434, 112 S. Ct. at 2063 (quoting Anderson v. Celebrezze, 460 U.S. 780, 789, 103 S. Ct. 1564, 1570, 75 L. Ed. 2d 547 (1983)). If the measure in question severely burdens the plaintiffs' rights, we apply strict scrutiny review. Burdick, 504 U.S. at 434, 112 S. Ct. at 2063. If, however, the law "imposes only 'reasonable, nondiscriminatory restrictions' upon the First and Fourteenth Amendment rights of voters, 'the State's important regulatory interests are generally sufficient to justify' the restrictions." Id. (quoting Anderson, 460 U.S. at 788, 103 S. Ct. at 1569).

The rights which the plaintiffs seek to vindicate in this case are the right to vote for the candidate of one's choice and the asserted right of an incumbent to again run for his or her office. Proposition 140's impact on these rights is not severe. As argued by the State, term limits on state officeholders is a neutral candidacy qualification, such as age or residence, which the State certainly has the right to impose. See Burdick, 504 U.S. at 433, 112 S. Ct. at 2063. With regard to incumbents, they may enjoy the incumbency of a single office for a number of years, and, as pointed out by the California Supreme Court, they are not precluded from running for some other state office.

Most important, the lifetime term limits do not constitute a discriminatory restriction. Proposition 140 makes no distinction on the basis of the content of protected expression, party affiliation, or inherently arbitrary factors such as race, religion, or gender. Nor does the Proposition "limit [ ] political participation by an identifiable political group whose members share a particular viewpoint, associational preference, or economic status." Anderson, 460 U.S. at 793, 103 S. Ct. at 1572.

Proposition 140's minimal impact on the plaintiffs' rights is justified by the State's legitimate interests. As the Proposition itself states, a lack of term limits may create "unfair incumbent advantages." Long-term entrenched legislators may obtain excessive power which, in turn, may discourage other qualified candidates from running for office or may provide the incumbent with an unfair advantage in winning reelection. As the Supreme Court stated in Thornton,¹ 

Term limits, like any other qualification for office, unquestionably restrict the ability of voters to vote for whom they wish. On the other hand, such limits may provide for the infusion of fresh ideas and new perspectives, and may decrease the likelihood that representatives will lose touch with their constituents.

Thornton, 514 U.S. at 837, 115 S. Ct. at 1871. California voters apparently perceived lifetime term limits for elected state officials as a means to promote democracy by opening up the political process and restoring competitive elections. This was their choice to make. Cf. Clements v. Fashing, 457 U.S. 957, 972, 102 S. Ct. 2836, 2848, 73 L. Ed. 2d 508 (1982).

We hold that Proposition 140's lifetime term limits do not violate the plaintiffs' first and fourteenth amendment rights. The judgment of the district court invalidating Proposition 140 is reversed and its injunction enjoining enforcement of the Proposition is vacated. The stay pending appeal is vacated as moot.

REVERSED.

O'SCANNLAIN, Circuit Judge, concurring in the result:

I concur in the result reached by Judge Thompson for the court and, except for a threshold concern, I would agree with his sound analysis of the merits. I also join in Judge Rymer's thoughtful discussion of the Rooker-Feldman doctrine and her analysis of res judicata. I write separately for two reasons: first, to suggest the utter absence of a federal question--irrespective of the Rooker-Feldman doctrine or res judicata--and hence an additional ground for our lack of jurisdiction to hear the case at all; and second, to respond to Judge Fletcher's dissent in which she urges that Proposition 140 is constitutionally infirm for lack of adequate "notice" to California voters.

* I have grave doubt that our court has jurisdiction to review this case, irrespective of Judge Rymer's concurring opinion or Judge Schroeder's dissent. There is compelling authority that a state's adoption of term limits on its own government's elected officials fails to raise a substantial federal question.

In Moore v. McCartney, 425 U.S. 946, 96 S. Ct. 1689, 48 L. Ed. 2d 190 (1976), the Supreme Court last had the opportunity to rule on term limits for state officials and it summarily dismissed, for want of a federal question, an appeal from the West Virginia Supreme Court upholding term limits on state executive officials. See State ex rel Maloney v. McCartney, 159 W.Va. 513, 519-20, 223 S.E.2d 607 (1976).

The ruling in Moore has the same effect as a summary affirmance and is fully binding on all lower courts. See Mandel v. Bradley, 432 U.S. 173, 176, 97 S. Ct. 2238, 2240, 53 L. Ed. 2d 199 (1977) (per curiam); Wright v. Lane County Dist. Ct., 647 F.2d 940, 941 (9th Cir. 1981). "Summary dismissals for want of a substantial federal question are decisions on the merits that bind the lower courts until subsequent decisions of the Supreme Court suggest otherwise." Wright, 647 F.2d at 941; see also Mandel, 432 U.S. at 176, 97 S. Ct. at 2240 (summary dismissals "prevent lower courts from coming to the opposite conclusions on the precise issues presented and necessarily decided by those actions.") The "precedential value of a dismissal for want of a substantial federal question extends beyond the facts of the particular case to all similar cases." Wright, 647 F.2d at 941.

Moore affirmed the Maloney decision which rejected an incumbent state officeholder's federal constitutional claim that consecutive term limits violated the Fourteenth Amendment to the Constitution. In his dissent in U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 925, 115 S. Ct. 1842, 1913, 131 L. Ed. 2d 881 (1995), Justice Thomas (joined by Chief Justice Rehnquist and Justices O'Connor and Scalia) shed some light on the significance of Moore. Thornton, of course, concerned the issue of whether states could limit the terms of members of the Congress of the United States chosen by the electorate of such states. The Supreme Court held that the Qualifications Clause of the Constitution, Article I, Section 5, prohibited the establishment of term limits by a state for its representatives and senators serving in the Congress of the United States. Id. at 831, 115 S. Ct. at 1868. Justice Thomas disagreed with the majority that such laws should be reviewed under the Qualifications Clause, and instead noted that "laws that allegedly have the purpose and effect of handicapping a particular class of candidates traditionally are reviewed under the First and Fourteenth Amendments.... Term limit measures have tended to survive such review without difficulty." Id. at 925, 115 S. Ct. at 1913.

Justice Thomas was referring to laws imposed by a state on its federal representatives, as was at issue in Thornton, not by a state on its own officeholders. However, in support of the proposition that state-imposed term limits on federal representatives have tended to survive review without difficulty, Justice Thomas cited Moore, explaining that the Supreme Court dismissed the appeal from Maloney, "on the ground that limits on the terms of state officeholders do not even raise a substantial federal question under the First and Fourteenth Amendments." Thornton, 514 U.S. at 925, 115 S. Ct. at 1913 (emphasis added). The Supreme Court's disposition of Maloney on this ground is not surprising in light of the Court's recognition of the fundamental interest of a state in determining the structure of its government. See Gregory v. Ashcroft, 501 U.S. 452, 460, 111 S. Ct. 2395, 2400, 115 L. Ed. 2d 410 (1991). What is noteworthy about this reference, however, is that, as relevant as Moore may have been with regard to federal officials, it is compelling as applied to state term limits. Indeed, we are presented today with the same question the Supreme Court dismissed in Moore for want of a substantial federal question: the constitutionality of limits on the terms of state officeholders.

For us to ignore the precedential value of Moore, we must necessarily decide that there is a constitutionally significant difference between executive and legislative term limits or between permanent and consecutive term limits. It appears from Thornton that these differences were not particularly persuasive to Justice Thomas and those joining him as he drew no such distinction in his characterization of the disposition of Moore. Further, the fact that the myriad of courts which have considered the various types of term limits have all reached the same result--that term limits on state officeholders are constitutional¹ --also leads me to believe that, if there is any difference, it is not of constitutional significance. Moreover, the California Supreme Court, when presented squarely with the argument that Maloney (and, by inference Moore) was of limited significance to Proposition 140's lifetime legislative term limits because it "involved a limitation on consecutive terms of a Governor," responded that "many, if not all of the considerations mentioned in Maloney (e.g., eliminating unfair incumbent advantages, dislodging entrenched political machines, restoring open access to the political process, and stimulating electorate participation) would apply with equal force to the legislative branch." Legislature of the State of California v. Eu, 54 Cal. 3d 492, 523-24, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991). Although I voice no opinion on the desirability of the policy considerations noted in Eu, I agree with the California Supreme Court that the considerations involved in executive term limits are equally applicable to legislative term limits, and the same holds for consecutive and lifetime term limits. I am not persuaded that there is a constitutionally significant distinction that would compel us to disregard Moore.

Judge Reinhardt, writing for the original panel in Jones v. Bates, 127 F.3d 839, 850 n. 13 (9th Cir. 1997), responded to the State's assertion that the constitutionality of Proposition 140 was controlled by Moore by noting that extensive intervening doctrinal developments counseled against continued reliance on that case, citing Hicks v. Miranda, 422 U.S. 332, 344, 95 S. Ct. 2281, 2289, 45 L. Ed. 2d 223 (1975), overruled by Mandel, 432 U.S. 173, 97 S. Ct. 2238, 53 L. Ed. 2d 199, for the proposition that "reliance on the fact that the Supreme Court has branded a particular question unsubstantial ceases to be appropriate 'when doctrinal developments indicate otherwise.' " In support of the contention that doctrinal developments indicate otherwise, Judge Reinhardt cited the Supreme Court's decisions in Burdick v. Takushi, 504 U.S. 428, 112 S. Ct. 2059, 119 L. Ed. 2d 245 (1992), Anderson v. Celebrezze, 460 U.S. 780, 103 S. Ct. 1564, 75 L. Ed. 2d 547 (1983), and Thornton, 514 U.S. 779, 115 S. Ct. 1842, 131 L. Ed. 2d 881. Upon review of these cases, however, along with the other relevant cases decided by the Supreme Court in the years between Moore and Bates, I am not persuaded that doctrinal developments have significantly changed Moore's fundamental premise. See Hicks, 422 U.S. at 344, 95 S. Ct. at 2289.

As an initial matter, Thornton does not speak at all to the question of whether limits on the terms of state officeholders raises a substantial federal question. Thornton is concerned exclusively with the constitutionality of state-imposed term limits on federal officeholders. The Supreme Court's decision was bottomed on the Qualifications Clause and the notion of a "national citizenship," neither of which, of course, has any relevance to this case. See Thornton, 514 U.S. at 806, 115 S. Ct. at 1856 (" [T]he Qualifications Clauses were intended to preclude the States from exercising any [control over congressional qualifications] and to fix as exclusive the qualifications in the Constitution."); see also Thornton, 514 U.S. at 841-844, 115 S. Ct. at 1873-74 (Kennedy, J., concurring) (discussing notion of a "national citizenship" and noting that "the National Government is and must be controlled by the people without collateral interference by the States.") Therefore, while Thornton would appear to the undiscerning eye to offer guidance on the issue we face today, upon closer examination, it is clear that it is considerably more limited in scope if not entirely inapposite.

As to other intervening doctrinal developments, although it is true that the Supreme Court has decided noteworthy ballot access cases in the period between Moore and Bates, it is doubtful that these cases detract from Moore 's precedential value. Significantly, in no case since Moore has the Supreme Court responded so directly to the issue before us today. While subsequent cases have presented numerous peripheral ballot access issues to the Court, none has presented the question of the constitutionality of state-imposed term limits on state officeholders as squarely as Moore. For this reason alone, I remain skeptical of the failure of our court to take this case into consideration in disposing of the matter before us.

Moreover, Moore is not inconsistent with any of the significant ballot access cases. As noted in Clements v. Fashing, 457 U.S. 957, 964, 102 S. Ct. 2836, 2844, 73 L. Ed. 2d 508 (1982), the Supreme Court has departed from "traditional equal protection analysis ... in two essentially separate, although similar, lines of ballot access cases," those which concern restrictions which are based on wealth, see Clements, 457 U.S. at 964, 102 S. Ct. at 2844, (citing Bullock v. Carter, 405 U.S. 134, 144, 92 S. Ct. 849, 856, 31 L. Ed. 2d 92 (1972); Lubin v. Panish, 415 U.S. 709, 717-18, 94 S. Ct. 1315, 1320-21, 39 L. Ed. 2d 702 (1974)), and those which impose burdens on new or small political parties or independent candidates. Id. (citing, inter alia, Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 99 S. Ct. 983, 59 L. Ed. 2d 230 (1979); Storer v. Brown, 415 U.S. 724, 94 S. Ct. 1274, 39 L. Ed. 2d 714 (1974); American Party of Texas v. White, 415 U.S. 767, 94 S. Ct. 1296, 39 L. Ed. 2d 744 (1974)). Burdick, 504 U.S. at 441-42, 112 S. Ct. at 2067-68 (approving Hawaii's ban against write-in candidates), Anderson, 460 U.S. at 806, 103 S. Ct. at 1579 (striking down state-imposed filing requirements on presidential candidates), and Timmons v. Twin Cities, --- U.S. ----, ----, 117 S. Ct. 1364, 1375, 137 L. Ed. 2d 589 (1997) (upholding state antifusion law prohibiting candidates from appearing on ballot as candidate of more than one political party), all clearly fit into the latter category. See Clements, 457 U.S. at 964, 102 S. Ct. at 2844. Term limits on state officials, however, do not fit into either category, and, more importantly for purposes of the Supreme Court's analysis, do not unfairly or unnecessarily burden "the availability of political opportunity." Id. (quoting Lubin, 415 U.S. at 716, 94 S. Ct. at 1320). While a failure to provide access to the ballot for new or small political parties or independent candidates or on the basis of wealth may effectively result in the exclusion of unique ideas and viewpoints from the political marketplace, the imposition of term limits raises no similar concerns. Term limits apply to politicians of every stripe regardless of party affiliation. It is understandable that the Supreme Court would consider the two lines of cases noted in Clements to raise a substantial federal question as they have a significant potential clearly to infringe upon First and Fourteenth Amendments rights. See e.g., Clements, 457 U.S. at 965, 102 S. Ct. at 2844 (burdensome requirements on small and independent parties "may burden First Amendment interests in ensuring freedom of association"). Terms limits, which have no disparate impact on any cognizable group, have less of a potential so to offend.

Consequently, I submit that intervening doctrinal developments in the ballot access cases have not weakened Moore 's message that term limits on state officeholders do not present a federal question. As Moore is the last word from the Supreme Court on the issue of term limits on state officeholders, we are compelled to give it appropriate weight in disposing of this case.

II

Judge Fletcher argues in her dissent that Proposition 140 is defective because the voters were not provided with "notice that they were voting on a severe limitation to their fundamental right to vote for candidates of their choice." See infra, Judge Fletcher dissenting opinion at 867; see also infra, Judge Fletcher dissenting opinion at 866 (incorporating Bates, 127 F.3d at 856-63). Although the court's opinion avoids this issue by assuming, without deciding, that a federal court may determine whether a state has given adequate "notice" to its voters in connection with a statewide initiative ballot measure, I feel that Judge Fletcher's analysis should not go unchallenged. First, I will examine whether the people have a "fundamental right to vote for candidates of their choice," and then I will consider whether some special "notice" is necessary when the voters of a state enact legislation via the initiative process.

* The assertion that the people have a "fundamental right to vote for candidates of their choice" implicates at least two distinct, though closely-related, rights: the right to vote and the right to run for elective office. See Lubin, 415 U.S. at 716, 94 S. Ct. at 1320 (" [V]oters can assert their preferences only through candidates or political parties.... The right of a party or an individual to a place on the ballot is entitled to protection and is intertwined with the rights of voters.") Consequently, in determining whether people have a fundamental right to vote for candidates of their choice, it is necessary to examine both the established right to vote and the asserted right to be a candidate and how the two interact.

Voting has long been recognized by the Supreme Court as a "fundamental right." See Harper v. Virginia Board of Elections, 383 U.S. 663, 667, 86 S. Ct. 1079, 1081, 16 L. Ed. 2d 169 (1965); Reynolds v. Sims, 377 U.S. 533, 561-62, 84 S. Ct. 1362, 1381-82, 12 L. Ed. 2d 506 (1964). It is axiomatic to say that the right to vote lies at the very core of our system of representative democracy. The right to vote in federal elections is conferred explicitly in Article I, Section 2 of the Constitution. See U.S. CONST. art. I, § 2. The Constitution is silent, however, on the right to vote in state elections; not surprisingly, it does not grant such a right, nor does it bestow upon Congress the power to determine the qualifications of state voters. See Harper, 383 U.S. at 667, 86 S. Ct. at 1081; see also Oregon v. Mitchell, 400 U.S. 112, 130, 91 S. Ct. 260, 267, 27 L. Ed. 2d 272 (1970). It is within the power of the several states to determine voter qualifications for state elections, so long as said qualifications do not violate any constitutional provision or "any restriction that Congress, acting pursuant to its constitutional powers, has imposed." Harper, 383 U.S. at 667, 86 S. Ct. at 1081 (quoting Lassiter v. Northampton Cty. Election Board, 360 U.S. 45, 51, 79 S. Ct. 985, 989, 3 L. Ed. 2d 1072 (1959)). As noted by Professor Tribe, every state imposes some restrictions on the franchise. See Laurence H. Tribe, American Constitutional Law § 13-10, at 1084 (2d ed.1988) (hereinafter, "Tribe"). States have imposed restrictions requiring that voters be of a certain reasonable age, that felons be denied the right to vote, see Richardson v. Ramirez, 418 U.S. 24, 56, 94 S. Ct. 2655, 2671, 41 L. Ed. 2d 551 (1974), and that voters be state citizens and residents. See, e.g., Marston v. Lewis, 410 U.S. 679, 679-80, 93 S. Ct. 1211, 1212, 35 L. Ed. 2d 627 (1973) (upholding 50-day voter residency requirement). These restrictions plainly infringe upon the right to vote--in some instances, completely prohibiting access to the franchise for millions²  of people--yet all have survived judicial scrutiny at the highest level. Consequently, although the right to vote is considered fundamental, fundamental is not commensurate with absolute.

The necessary corollary to the notion that the people have a fundamental right to vote for candidates of their choice is that there exists a constitutional right to vote for a particular candidate. It seems beyond cavil that the former cannot survive without the latter. The right to vote for a particular candidate, in turn, perforce implicates the right to stand for elective office. The district court in Bates v. Jones, 958 F. Supp. 1446, 1460 (N.D.Ca.1997), in identifying the constitutional rights implicated by Proposition 140's lifetime legislative term limits, gleaned from the Supreme Court's decision in Burdick, 504 U.S. at 441, 112 S. Ct. at 2067 a "right to vote for [a] particular candidate." The district court proclaimed that voters can be denied the opportunity to vote for a particular candidate only if "constitutionally adequate" means existed for that candidate to appear on the ballot. See Bates, 958 F. Supp. at 1460. I must respectfully disagree. The district court reads far too much into Burdick 's discussion of the means of gaining access to the Hawaii ballot. Indeed, it is well-established that " [a]lthough [a voter] is guaranteed an equal voice in the election of those who govern, [he or she] does not have an unlimited right to vote for any particular candidate." Burdick v. Takushi, 937 F.2d 415, 418 (9th Cir. 1991), aff'd, 504 U.S. 428, 112 S. Ct. 2059, 119 L. Ed. 2d 245 (1992). As we noted in Burdick, the Supreme Court has upheld numerous restrictions imposed by the states on who may run for certain state offices. Id.; see e.g., Munro v. Socialist Workers Party, 479 U.S. 189, 196-97, 107 S. Ct. 533, 537-38, 93 L. Ed. 2d 499 (1986) (upholding state requirement that minor-party candidates receive at least 1% of votes in state's multi-party primary election for candidate to appear on ballot); Clements, 457 U.S. at 972, 102 S. Ct. at 2848 (upholding provisions of state constitution which provided for automatic resignation of certain state and local officeholders who became candidates for any other state office and which prevented most government officials from becoming eligible to serve in state legislature before completing current term of office); Storer, 415 U.S. at 728-29, 94 S. Ct. at 1278-79 (upholding state disaffiliation provision which prevented person from running as independent candidate if he registered with political party during year prior to immediately preceding primary); American Party of Texas, 415 U.S. at 782, 94 S. Ct. at 1306 (state can deny place on ballot to frivolous candidate by requiring all candidates to "demonstrate a significant, measurable quantum of community support.") Consequently, while there is a right to equal participation in voting, a constitutional right to vote for a particular candidate simply does not exist.³ 

Further, if there is a right to vote for a particular candidate, such right necessarily implicates a corresponding constitutional right to be a candidate. Judge Fletcher suggests in her dissent that "every attempt by a state to set restrictive qualifications concerning who can and who cannot serve as a candidate for elective office implicates rights that are 'fundamental' and requires careful scrutiny." See infra, Judge Fletcher dissenting opinion at 868. The district court also suggested that there is a "right to run for a particular office" that requires protection akin to that afforded the right to vote. See Bates, 958 F. Supp. at 1461 (citing Lubin, 415 U.S. at 716, 94 S. Ct. at 1320). Once again, I respectfully disagree both with Judge Fletcher and the district court. The interest of an individual in being a candidate is of no exceptional constitutional significance; it does not merit a departure from traditional equal protection principles. "Far from recognizing candidacy as a 'fundamental right,' we have held that the existence of barriers to a candidate's access 'does not of itself compel close scrutiny.' " Clements, 457 U.S. at 962, 102 S. Ct. at 2843 (quoting Bullock, 405 U.S. at 143, 92 S. Ct. at 855).

For the foregoing reasons, the asserted constitutional right of the people "to vote for the candidates of their choice" simply does not exist. Consequently, it is evident that no "fundamental right" was "severely limited" by Proposition 140 and thus, the dissent's analysis is fatally flawed.

B

As to the contention that Proposition 140's lifetime legislative term limits must fail because the voters of California were not provided with adequate "notice" of the "severe limitation" the provision would impose on their fundamental rights, there is even less foundation in law or tradition. While I agree with the court's opinion that sufficient "notice" of the extent of the term limits provision was provided to the voters of the State of California, I hesitate to assume for even a moment that such "notice" was required or even within the province of a federal court to determine.⁴  See, supra majority opinion at 845-846.

Judge Reinhardt, writing for the original panel, held that, although the proper interpretation of Proposition 140 is purely a matter of state law, "the issue whether the people's fundamental political rights may be severely burdened by means of an 'ambiguous' initiative that 'the average voter' would only 'likely' understand is ... a question of federal law." Bates, 127 F.3d at 857. Judge Fletcher argues in her dissent that, while the California Supreme Court in Eu was only concerned with what the "average" voter "likely" believed about Proposition 140, "surely the adequacy of notice cannot turn on what a person of average intelligence 'likely' understood." See infra, Judge Fletcher dissenting opinion at 866. According to Judge Fletcher, evidently, the federal standard for adequate "notice" when fundamental rights are at stake requires a different and more stringent inquiry than the state standard. Searching the Constitution, however, I am unable to locate an "ignorant voter clause" that vests federal courts with the power to review voter-enacted legislation to ensure that enough people were capable of understanding what they voted for at the ballot. In all fairness, Judge Fletcher does not attempt to ground this newly-discovered right in the Constitution. Instead, it appears to be simply yet another novel right that "implicates due process concerns," see Bates, 127 F.3d at 857, created by weaving together snippets from various cases,⁵  none of which would seem to betray that it would one day have the dubious distinction of being called upon for this purpose.

Such "notice" requirement is most likely grounded in an antipathy for and distrust of the initiative process which finds support in neither the Constitution nor Supreme Court precedent. Although it has been argued that the Supreme Court reviews popular legislation more closely than legislation enacted by representatives, see Julian N. Eule, Judicial Review of Direct Democracy, 99 Yale L.J. 1503, 1562-65 (1990); Mark Slonim & James H. Lowe, Comment, Judicial Review of Laws Enacted by Popular Vote, 55 Wash. L.Rev. 175, 194-96 (1979), an examination of the case law does not bear out such assertion. As noted in City of Eastlake v. Forest City Enterprises, Inc., 426 U.S. 668, 672-73, 96 S. Ct. 2358, 2361-62, 49 L. Ed. 2d 132 (1976), " [i]n establishing legislative bodies, the people can reserve to themselves power to deal directly with matters which might otherwise be assigned to the legislature." The Supreme Court applies the very same standard of review both to popular and to representative legislation and has held that the fact that legislation was enacted by initiative "is without constitutional significance."⁶  Lucas v. Forty-Fourth General Assembly, 377 U.S. 713, 737, 84 S. Ct. 1459, 1474, 12 L. Ed. 2d 632 (1964) (striking down, on equal protection grounds, apportionment scheme that violated "one person-one vote" principle notwithstanding approval by voters of state).

To engraft a due process "notice" requirement on popular legislation is to invite the federal courts to look behind initiatives to ensure that voters were capable of understanding the potential consequences of their actions and acted accordingly. This is simply beyond our province. The Supreme Court has historically presumed that legislators are aware of the consequences of the laws which they enact and has declined to invalidate legislation on the ground that "Congress was unaware of what it accomplished or ... was misled by the groups that appeared before it." See United States Railroad Retirement Bd. v. Fritz, 449 U.S. 166, 179, 101 S. Ct. 453, 461, 66 L. Ed. 2d 368 (1980) ("The language of the statute was clear and we have historically assumed that Congress intended what it enacted."); see also Taxpayers to Limit Campaign Spending v. Fair Political Practices Comm'n, 51 Cal. 3d 744, 274 Cal. Rptr. 787, 801, 799 P.2d 1220, 1235 (1990) ("In order to further the fundamental right of the electorate to enact legislation through the initiative process, this court must on occasion indulge in a presumption that the voters thoroughly study and understand the content of complex initiative measures.") Further, it has long been established that there is no due process right to "notice" in the enactment of legislation. See Bi-Metallic Investment Co. v. State Board of Equalization, 239 U.S. 441, 445, 36 S. Ct. 141, 142, 60 L. Ed. 372 (1915) (no due process right to notice before government acts in legislative capacity). In light of the fact that there is no constitutionally significant difference between popular legislation and representative legislation, see Lucas, 377 U.S. at 737, 84 S. Ct. at 1474, we must reject the temptation to create a due process right to "notice" in the initiative process. We simply have no franchise to presume that the citizens of California could not understand what they were doing at the ballot box on November 6, 1990.

III

Like Judge Thompson, I would not interfere with the will of the people of California in adopting Proposition 140; I would reach such result, however, by the route charted by Judge Rymer or Part I of the foregoing analysis. With respect, for the reasons expressed in Part II, I must fundamentally reject the view expressed by Judge Fletcher that federal courts have a writ to impose a newly-minted "notice" requirement on the voters of California.

RYMER, Circuit Judge, with whom O'SCANNLAIN, Circuit Judge, joins, concurring in the result:

I agree with Judge Thompson's disposition of Bates's appeal, assuming the merits are reached and that an Anderson /Burdick balancing test is required. However, I write separately for two reasons: First, to explain why I believe that the district court should not have entertained this action because it involves the same claim by related parties already considered and decided by the California Supreme Court. In effect, we are asked to review the state supreme court's opinion that Proposition 140 is constitutional--but that's not our right to do. Only the United States Supreme Court can overturn a decision by the highest court of a state. For a federal district court to make an independent determination of constitutionality in the face of a final state court adjudication on the same point implicates well-settled doctrines of finality, federalism, and the limits of appellate jurisdiction. Therefore, in my view, the district court lacked the authority to proceed; and in any event, should have declined to do so on grounds of res judicata.

Second, assuming that we have jurisdiction, I write separately to suggest that the state's interest in Proposition 140 is somewhat different, and even stronger than Judge Thompson acknowledges, because the people of the State of California exercised their own constitutional right to choose the form of their representation in the legislative branch of state government. In this they acted in accordance with the Constitution's guarantee that citizens of each state shall have the right to determine the structure of their own government so long as it is republican in form. Since the state's interest in structuring the state legislature is firmly rooted in the Constitution, a decision to adopt term limits for state officers is presumptively constitutional. In that Proposition 140 is content neutral and no one suggests it discriminates on any basis plainly precluded by other provisions of the Constitution, the presumption is not overcome. For the same reasons, the state's interest in prescribing the qualifications of state officeholders is so powerful, given its constitutional dimension, that the decision to set the number of terms in office as a qualification easily survives whatever scrutiny is required.

* In Legislature v. Eu, 54 Cal. 3d 492, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991), the Legislature, several individual legislators, and various citizens, voters and taxpayers brought the same constitutional challenges to Proposition 140 that Bates (then, as now, a member of the Legislature), other legislators, and other voters bring in this action. The Bates plaintiffs are represented by the same law firm that represented the plaintiffs in Eu. Apart from budgetary limitations in Proposition 140 that were at issue there but not here, the exact challenges to the constitutional validity of Proposition 140 were raised, considered, and decided by the California Supreme Court in Eu. The plaintiffs in both cases claim that the lifetime ban substantially burdens their fundamental First and Fourteenth Amendment rights. The court considered and balanced, under Anderson, the character and extent of injury to the incumbent's right to run for public office and the voters' right to reelect the incumbent; the legitimacy of the state's asserted interests (restoring competitive elections, encouraging qualified candidates to seek public office, and eliminating unfair incumbent advantages); and the necessity of imposing the lifetime restriction. Having done so, the California Supreme Court concluded that the interests of the state in incumbency reform outweigh any injury to incumbent office holders and those who would vote for them, and therefore held that Proposition 140 passes constitutional muster.

How can this ruling possibly be revisited?

* No federal court except for the United States Supreme Court has appellate jurisdiction over final decisions of the California Supreme Court. As the United States Supreme Court made clear in Rooker v. Fidelity Trust Co., 263 U.S. 413, 44 S. Ct. 149, 68 L. Ed. 362 (1923), when a state court decides a federal constitutional question that actually arose in the cause before it, its decision is open to reversal or modification by the United States Supreme Court. Unless and until that happens, however, it is an effective and conclusive adjudication. See also District of Columbia Court of Appeals v. Feldman, 460 U.S. 462, 103 S. Ct. 1303, 75 L. Ed. 2d 206 (1983). In Eu, the California Supreme Court had to decide whether term limits imposed by Proposition 140 violated the United States Constitution, and its decision, whether right or wrong, was a proper exercise of that court's jurisdiction. Accordingly, no court of the United States other than the Supreme Court can entertain a proceeding to reverse or modify it because to do so would be to exercise appellate jurisdiction which federal district courts (and we) do not have over state court judgments.

As we explained in Dubinka v. Judges of the Superior Court, 23 F.3d 218 (9th Cir. 1994):

Federal district courts may exercise only original jurisdiction; they may not exercise appellate jurisdiction over state court decisions. This rule arises from the interplay of two jurisdictional statutes: 28 U.S.C. § 1331, which grants district courts original jurisdiction over "civil actions arising under" federal law, and 28 U.S.C. § 1257, which grants the Supreme Court the right to review "final judgments ... rendered by the highest court of a State."

Id. at 221 (citations omitted); see also Atlantic Coast Line R.R. Co. v. Brotherhood of Locomotive Eng'rs, 398 U.S. 281, 90 S. Ct. 1739, 26 L. Ed. 2d 234 (1970).

The "Rooker-Feldman " doctrine applies even when the challenge to a state court decision involves federal constitutional issues, because state courts are as competent as federal courts to decide federal constitutional issues. Dubinka, 23 F.3d at 221; Worldwide Church of God v. McNair, 805 F.2d 888, 891 (9th Cir. 1986). Any other rule would result in a waste of judicial resources, and cause needless friction between state and federal courts--as this case amply illustrates. See McNair, 805 F.2d at 891. Indeed, to re-review the constitutionality of Proposition 140 put the district court on a collision course with the California Supreme Court. At the very least this is unseemly, because the California Supreme Court had the right to decide Proposition 140's constitutionality. The United States Supreme Court saw fit to let that court's ruling stand. Appellate authority would be turned upside down if the federal district court were able effectively to "overrule" the California Supreme Court. Moreover, as a practical matter, disagreement between a federal district court (or this court) and the California Supreme Court would make a mess of California elections. This is why Rooker-Feldman instructs that a federal district court lacks jurisdiction to start down this path.

Of course, federal district courts do have jurisdiction over a general constitutional challenge to a state enactment that does not require review of a final state court decision in a particular case.¹  See, e.g., Feldman, 460 U.S. at 482-86, 103 S. Ct. at 1314-15 (challenge to application of bar admission rule differed from challenge to its constitutionality). In a close case, we decide whether a case falls on the side of a permissible general challenge, or the side of an impermissible appeal, by asking whether the constitutional challenge raised in the federal court was "inextricably intertwined" with the state court's decision. See, e.g., McNair, 805 F.2d at 890-93 (holding that we lacked jurisdiction); Dubinka, 23 F.3d at 221-22 (applying test but holding that we had jurisdiction). However, here, there can be no serious question that the constitutional challenge was "inextricably intertwined"; it is, in fact, the only thing decided by the California Supreme Court in Eu and it is the only thing the federal district court was asked to decide in Bates.

It seems clear to me, therefore, that the district court lacked jurisdiction and (apart from saying that), so do we.

B

But even if we have jurisdiction, related principles of finality embedded in res judicata also counsel against our revisiting what the California Supreme Court decided. Unsurprisingly, under California law, which we apply to this question,²  the doctrine of res judicata precludes parties or their privies from relitigating a cause of action or issue that has been finally determined by a court of competent jurisdiction. Bernhard v. Bank of America, 19 Cal. 2d 807, 810, 122 P.2d 892 (1942); Lucido v. Superior Ct., 51 Cal. 3d 335, 341, 272 Cal. Rptr. 767, 795 P.2d 1223 (1990) (party asserting bar must show the factual or legal issues are identical; the issue was actually litigated and there was a final judgment on the merits; and the party to be estopped was in privity). California recognizes, as we do, that claim and issue preclusion curtail vexatious litigation, promote judicial economy, and avoid the issuance of inconsistent judgments that undermine the integrity of the judicial system. See Lynch v. Glass, 44 Cal. App. 3d 943, 948, 119 Cal. Rptr. 139 (1975). Each supports the state's position in this case.

The legal issue in Eu and Bates is identical. The issue was actually, and actively, litigated in the California Supreme Court, which rendered a final judgment on the merits. And legislators as well as voters were parties to Eu, as they are in Bates. To be sure, they are different legislators, and different constituents, but I fail to see what difference that makes because their respective interests (in succeeding themselves or voting for incumbents) are precisely the same. So is their lawyer. Thus, the Bates plaintiffs share an identity of interest with, and adequate representation by, the losing parties in Eu. Under these circumstances, I cannot believe that any court in California would not feel itself, and a new set of voters or legislators, bound by the California Supreme Court's decision in Eu. See Dyson v. California State Personnel Bd., 213 Cal. App. 3d 711, 723-24, 262 Cal. Rptr. 112 (1989) (recognizing that the question is essentially one of policy, focusing on whether the relationship between the party bringing the earlier suit and the nonparty bringing the current suit is "sufficiently close" to warrant preclusion).³ 

While courts in California may decline to apply the doctrine of collateral estoppel when preclusion would not serve the public interest and might work an injustice, see e.g., Kopp v. Fair Political Practices Comm'n, 11 Cal. 4th 607, 47 Cal. Rptr. 2d 108, 905 P.2d 1248 (1995), City of Sacramento v. State, 50 Cal. 3d 51, 266 Cal. Rptr. 139, 785 P.2d 522 (1990); Louis Stores, Inc. v. Dep't of Alcoholic Beverage Control, 57 Cal. 2d 749, 22 Cal. Rptr. 14, 371 P.2d 758 (1962), neither reason would require a California court (or us) to revisit the constitutionality of Proposition 140. Nothing substantive changed about the application of Proposition 140 from October 10, 1991 (when Eu was decided), or from March 9, 1992 (when certiorari was denied), until 1995, when Bates filed this action--except that Bates himself came up against the term limits ban.⁴  That, however, cannot suffice, for the issue of constitutionality which Bates now raises was raised by his colleagues, and was fully litigated, in Eu by the state's highest tribunal. Likewise, the same constitutional issues that Bates's constituents raise were raised by his colleagues' constituents in Eu. Therefore, no injustice inheres for new or different voters, or for new or different legislators; they are equally stuck with the ban on term limits. Nor do the intervening decisions in Burdick and Thornton suggest otherwise: assuming the Anderson test applies, Burdick supports the decision reached by the California Supreme Court (which applied the Anderson standard) because Burdick simply affirmed that the Anderson analysis also governs inquiries into the propriety of state election laws; and Thornton involved the attempt by a state to impose term limits on a federal office, which raises different constitutional questions from the state's adopting term limits on state offices.

Indeed, to revisit the constitutionality of Proposition 140 subverts, rather than serves, the public interest. The California Supreme Court (the final arbiter of state law, and a court quite competent to decide federal constitutional law), made its decision and that decision was left in place when the United States Supreme Court refused to grant a writ of certiorari. It is obviously important for California elections to be held without continual confusion over the governing law. That is a powerful reason the validity of Proposition 140 should be settled by Eu.

I realize that, at the end of the day, whether preclusion principles apply is a matter of policy. But I believe that California courts would conclude that there are no "exceptional circumstances" which justify applying the "extremely narrow" public interest exception in this case. Arcadia Unified Sch. Dist. v. State Dep't of Educ., 2 Cal. 4th 251, 259, 5 Cal. Rptr. 2d 545, 825 P.2d 438 (1992).

II

As Professor Tribe has pointed out, decisions about how people are to participate in their own representative government and how they choose to represent themselves is "the very essence of all self-government." Laurence H. Tribe, American Constitutional Law 398 (2d ed.1988). So, even assuming that the district court had jurisdiction and properly refused to preclude repetitive litigation on the constitutionality of Proposition 140, I start (and mostly stop) with the constitutional right exercised by the people of the State of California to have themselves represented in the legislative branch of their state government by citizen-legislators. That right resides in Article IV, § 4 of the United States Constitution, which declares that " [t]he United States shall guarantee to every State in this Union a Republican Form of Government."

As Professor Merritt has explained, the Guarantee Clause restricts the federal government's power to interfere with the organizational structure and governmental processes chosen by a state's residents.... In order to ensure that state and local government remain responsive to their constituents, ... citizens must have the power to choose the governmental forms that work best for them. The guarantee clause, therefore, grants states control over their internal governmental machinery.

Deborah Jones Merritt, The Guarantee Clause and State Autonomy: Federalism for a Third Century, 88 Colum. L. Rev. 1, 41 (1988). Although there is much debate about the full meaning of Article IV, § 4,⁵  at a minimum it promises the states authority to set qualifications for state and local offices (so long, at least, as they do not otherwise discriminate on an invidious basis plainly prohibited by the Constitution). See Deborah Jones Merritt, Republican Governments and Autonomous States: A New Role for the Guarantee Clause, 65 U. Colo. L.Rev. 815, 828-829 (1994).

This principle has long been recognized by the Supreme Court. As the Court wrote in In re Duncan, 139 U.S. 449, 461, 11 S. Ct. 573, 577, 35 L. Ed. 219 (1891), "the distinguishing feature" of a republican form of government "is the right of the people to choose their own officers for governmental administration, and pass their own laws." Most recently, Justice O'Connor reiterated the point writing for the plurality in Gregory v. Ashcroft, 501 U.S. 452, 111 S. Ct. 2395, 115 L. Ed. 2d 410 (1991):

The present case concerns a state constitutional provision through which the people of Missouri establish a qualification for those who sit as their judges. This provision goes beyond an area traditionally regulated by the States; it is a decision of the most fundamental sort for a sovereign entity. Through the structure of its government, and the character of those who exercise government authority, a State defines itself as a sovereign. "It is obviously essential to the independence of the States, and to their peace and tranquility, that their power to prescribe the qualifications of their own officers ... should be exclusive, and free from external interference, except so far as plainly provided by the Constitution of the United States." Taylor v. Beckham, 178 U.S. 548, 570-571, 20 S. Ct. 890, 897-898, 44 L. Ed. 1187 (1900). See also Boyd v. Nebraska ex rel. Thayer, 143 U.S. 135, 161, 12 S. Ct. 375, 381, 36 L. Ed. 103 (1892) ("Each State has the power to prescribe the qualifications of its officers and the manner in which they shall be chosen.").

Gregory, 501 U.S. at 460, 111 S. Ct. at 2400. The authority of the people of the states to determine the qualifications of their officials

is an authority that lies at " 'the heart of representative government.' " It is a power reserved to the States under the Tenth Amendment and guaranteed them by that provision of the Constitution under which the United States "guarantee [s] to every State in this Union a Republican Form of Government."

Id. at 463, 111 S. Ct. at 2402 (citations omitted).

Accordingly, when the people of California enacted Proposition 140 to set the number of terms in office as a qualification for members of their legislature, they were exercising an authority "that lies at the heart of representative government." As the authority to determine the terms upon which state offices shall be held is grounded in Article IV, § 4, and to define how they want to be represented is a right secured to the citizens of each state, a term limits qualification must be presumptively constitutional. Therefore, unless the qualification is plainly prohibited by some other provision in the Constitution, it is constitutional.

Proposition 140 discriminates against no one, applies equally to all citizens, and does nothing to restrict the franchise or freedom of association among like-minded voters and public officeholders (past, present or future). Thus, it is constitutional, for it is well within the range of decisions guaranteed to the citizens of California without federal interference.

For these reasons I would not start by analyzing Proposition 140 under the Anderson/Burdick test, because term limits are a qualification for office--not for access to the ballot. Yet I agree that if that test applies, California's lifetime term limits pass muster for the reasons explained by Judge Thompson and one more: the state's interest in the structure of its legislature is among the strongest possible interests that the citizens of California could have, because it lies at the core of the state's constitutional authority.

HAWKINS, Circuit Judge, concurring:

I join the majority opinion because, although I believe there are responsible arguments to be made on both sides, I do not believe the Constitution of the United States restrains citizens of a state from imposing term limits, even a lifetime ban on future service, on their elected state representatives. Our role here, of course, is to judge not the wisdom of such a measure, but its constitutionality.

I also write separately to express briefly two other points that deserve mention.

First, while I agree with the majority in its res judicata analysis, I would go further than the majority. Not only does the public interest exception enable us to review this matter, but even if the exception did not apply, this case would not be barred by res judicata because the parties in this case are not in complete privity with those in Legislature v. Eu, 54 Cal. 3d 492, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991), cert. denied, 503 U.S. 919, 112 S. Ct. 1292, 117 L. Ed. 2d 516 (1992).

The interests of Bates (and the other named plaintiffs) could not have been adequately represented by the Legislature or the particular legislators in Eu. Bates et al. assert their own individual right to run for an additional term or, in the case of their supporters, to vote for the particular candidate of their choice. This right is not subsidiary to any interest the Legislature as an institution might have had in Eu. This raises real doubts about whether the Legislature or the particular legislators in Eu "might fairly be treated as acting in a representative capacity," Lynch v. Glass, 44 Cal. App. 3d 943, 119 Cal. Rptr. 139 (1975), and whether privity exists between the parties in the two actions.

Second, although I agree with the majority that "the relevant ballot materials and the surrounding context provided sufficient notice making it clear that Proposition 140 required lifetime bans," in the interest of balance, the record should be clear that the proponents of Proposition 140 have not exhibited complete consistency on this point. In fact, they argued for the opposite conclusion before the California Supreme Court.

SCHROEDER, Circuit Judge, dissenting in part:

I agree that the district court must be reversed and the injunction vacated. I respectfully disagree with the majority's holding that we should reach the merits of the constitutional controversy. The California Supreme Court has already decided it. Legislature v. Eu, 54 Cal. 3d 492, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991). The majority correctly recognizes that res judicata would ordinarily bar relitigation of such issues, but incorrectly concludes that the California Supreme Court would apply a "public interest" exception to permit relitigation.

Under the Full Faith and Credit Act, we must give the state court judgment the same preclusive effect that the California Supreme Court would afford it. U.S. Const. art. IV, § 1; 28 U.S.C. § 1738; Migra v. Warren City School Dist. Bd. of Educ., 465 U.S. 75, 104 S. Ct. 892, 79 L. Ed. 2d 56 (1984). There is nothing in California law to suggest the California Supreme Court would decide to revisit the same case it decided a few years ago, particularly when no intervening federal cases would compel it to change its analysis.

The majority cites Kopp v. Fair Political Practices Comm'n, 11 Cal. 4th 607, 47 Cal. Rptr. 2d 108, 115, 905 P.2d 1248, 1256 (1995). There, the California court applied its public interest exception to consider in the first instance whether to reform a state statute that a federal court had held unconstitutional. There had been no prior state litigation. Thus, California had a considerable public interest. There is no similar public interest to be served in this case by relitigating the identical claims that the same state court earlier decided.

The panel majority circumvented res judicata by a different route. It held that the particular legislators and voters who are plaintiffs in this case were neither parties to the state court litigation, nor in privity with those parties. The majority in my view correctly, albeit implicitly, recognizes that the panel's position is not sound. Identical voter and legislator interests were asserted by the parties in the earlier litigation, and the plaintiffs in both actions were represented by the same counsel. I believe the California courts would hold that the Legislature and voters in Eu acted in a representative capacity for the plaintiffs in this case, and that they are therefore privies barred by the doctrine of res judicata. See Bernhard v. Bank of America Nat'l Trust & Sav. Ass'n, 19 Cal. 2d 807, 122 P.2d 892 (1942).

I would reverse and remand for dismissal of the federal action. I would not decide the merits.

FLETCHER, Circuit Judge, with whom Circuit Judge PREGERSON joins, Concurring in Part and Dissenting in Part:

The majority today decides questions of great importance: May a State, consistent with the Constitution, impose a lifetime term limitation upon elected officials for state office, and, if so, how may the State go about enacting such a restriction? These are questions of first impression among the federal courts of appeal. With regret, I must conclude that the majority has not given them the careful consideration that they deserve.

The State of California argues, as do Judge Rymer and Judge Schroeder, that the question of the constitutionality of Proposition 140 is res judicata as between the parties to this action, and that we should not reach the merits of the plaintiffs' claims. It is, of course, well established that federal courts must give full faith and credit to a valid judgment rendered by a state court of competent jurisdiction. See 28 U.S.C. § 1738. More precisely, a federal court must accord such a judgment the same preclusive effect that it would receive in the rendering state. See Matsushita Elec. Indus. Co. v. Epstein, 516 U.S. 367, ----, 116 S. Ct. 873, 877, 134 L. Ed. 2d 6 (1996). California and Judges Rymer and Schroeder argue that the parties to the present action are the same parties that challenged the constitutionality of Proposition 140 in Legislature v. Eu, 54 Cal. 3d 492, 286 Cal. Rptr. 283, 816 P.2d 1309 (1991), or at least that they were in privity with the plaintiffs in that case, and hence that they should be bound by the judgment of the California Supreme Court, and their present claims should be barred. Without deciding whether Bates and the other plaintiffs are actually bound by the Eu decision, the majority concludes (as did the three-judge panel, in an alternative holding on the question) that California would apply its "public interest" exception to res judicata if this case were brought in state court, and so res judicata is no bar to the present action. See supra, majority opinion at 845-846; Jones v. Bates, 127 F.3d 839, 850-51 (9th Cir. 1997). I continue also to subscribe to the panel majority's opinion that these plaintiffs are not barred by res judicata because they were not parties to the prior state court proceeding. See id. at 848-49.

This leads me to respond to Judge Rymer's disagreement on this point contained in her separate concurrence on this issue. Judge Rymer invokes the Rooker-Feldman doctrine, see District of Columbia Court of Appeals v. Feldman, 460 U.S. 462, 103 S. Ct. 1303, 75 L. Ed. 2d 206 (1983); Rooker v. Fidelity Trust Co., 263 U.S. 413, 44 S. Ct. 149, 68 L. Ed. 362 (1923), in order to argue that the district court had no jurisdiction even to hear Bates' challenge to Proposition 140. In doing so, she apparently argues that, when the highest court of a state issues a ruling on the constitutionality of a state statute, the RookerFeldman doctrine prevents lower federal courts from ever conducting an independent review of the statute's constitutionality, regardless of whether the parties before the federal court are different from the parties in the prior state court proceeding. The majority properly declines to embrace this extraordinary proposition, which constitutes a serious misreading of the Rooker-Feldman doctrine.¹ A.

As Judge Rymer correctly explains, the Rooker-Feldman doctrine holds that federal courts have no jurisdiction to engage in appellate review of the merits of state court judgments. With some exceptions that are not relevant here, the jurisdiction of federal district courts is original, not appellate. See 28 U.S.C. § 1331. Congress has reserved the duty to hear appeals from individual state court judgments, even on issues of federal constitutional law, exclusively to the U.S. Supreme Court in our judicial system. See 28 U.S.C. § 1257; Dubinka v. Judges of the Superior Court, 23 F.3d 218, 221 (9th Cir. 1994).

Because it is a doctrine concerning the integrity of individual judgments, Rooker-Feldman, like the doctrine of res judicata, is applicable only when the parties in a second action were also parties, or in privity with parties, in a previous state court proceeding. It is well established that "one is not bound by a judgment in personam in a litigation in which [one] is not designated as a party," unless it is through a privity relationship or a certified representative. Richards v. Jefferson Cty., Ala., 517 U.S. 793, ----, 116 S. Ct. 1761, 1766, 135 L. Ed. 2d 76 (1996). Someone who is not a party to an action cannot, of course, request appellate review of that action. See, e.g., Fed. R.App. Proc. 3(c) ("A notice of appeal must specify the party or parties taking the appeal.") By the same token, someone who was not a party to a state court action cannot be charged with attempting to seek "appellate review" of that action, within the meaning of RookerFeldman, by bringing a separate action in federal court that happens to address issues that are similar or identical to the issues decided in the state court proceedings. The Rooker-Feldman doctrine prevents federal courts from reviewing state court judgments; it does not prevent federal courts from deciding issues that happen previously to have been decided in unrelated state court proceedings.

One might well ask what the difference is between the Rooker-Feldman doctrine and traditional principles of res judicata. In many cases, the answer will be that the difference is a purely formal one. As Judge Easterbrook has recently written, "a judgment that is not entitled to full faith and credit does not acquire extra force via the Rooker-Feldman doctrine." Kamilewicz v. Bank of Boston Corp., 100 F.3d 1348, 1350 (7th Cir. 1996) (Easterbrook, J., dissenting from denial of rehearing en banc). This observation holds true both in cases where a party claims that the state proceeding that produced a challenged judgment suffered from a constitutional infirmity (the type of claim that Judge Easterbrook was analyzing in Kamilewicz) and in cases where a party claims that the state court judgment should not bar her action in federal court because it would not bar the same action in a court of the rendering state. Thus, in most cases, the difference between the Rooker-Feldman doctrine and the principle of res judicata amounts to the difference between asking a federal court to review the merits of a binding state court judgment and asking a federal court to address the same issues de novo without reference to the prior judgment. If a party comes to federal court and asks to be relieved of a state court judgment because it was wrongly decided, Rooker-Feldman directs the district court to respond that it has no jurisdiction to engage in such review. If the same party comes into federal court and simply pleads the same claims that were decided in the state court judgment, without making reference to that judgment or asking the district court to review its merits, the Full Faith and Credit Act would require the federal courts to recognize the affirmative defense of res judicata when a defendant interposes the state court judgment as a bar to the plaintiff's claims.² 

It is difficult to articulate a general rule for identifying the circumstances under which the applicability of Rooker-Feldman and of res judicata are not essentially coextensive. This may be what inspired a leading commentator in the field of Civil Procedure to describe the Rooker-Feldman doctrine as "somewhat peculiar." Marcel Kahan & Linda Silberman, Matsushita and Beyond: The Role of State Courts in Class Actions Involving Exclusive Federal Claims, 1996 Sup. Ct. Rev. 219, 283 n.122. Perhaps the best explanation of the Rooker-Feldman doctrine is to be found by examining the highly unusual interplay between state and federal courts that obtained in the Feldman case itself.

Feldman arose out of the refusal by the District of Columbia Court of Appeals to admit Feldman to the practice of law in that jurisdiction. The Court of Appeals in the District had a blanket rule against admitting individuals, like Feldman, who had not attended an accredited law school, and it refused to make an exception in Feldman's case. See Feldman, 460 U.S. at 465-70, 103 S. Ct. at 1305-08. The "judgment" that Feldman challenged in federal district court was the refusal by the state court³  to grant him an exception and admit him to the D.C. bar. On review, the Supreme Court held that the denial of Feldman's petition did in fact constitute a "judgment" within the meaning of 28 U.S.C. § 1257, and hence that review of that denial could be had, if at all, only in the U.S. Supreme Court. See id. at 479-82, 486-88, 103 S. Ct. at 1313-14, 1316-17.

What is unusual about Feldman is that the claim of right in that case--the denial to a particular applicant of admission to a state bar--involved a matter of policy that was under the exclusive control of the court. A claimant has occasion to challenge his denial to the bar of a state only after denial of admission by the state court itself. Since the Court found in Feldman that a denial of admission to the bar constitutes a "judicial proceeding," id. at 479-82, 103 S. Ct. at 1313-14 such denials form a unique class of cases in which every claimant will always already be saddled with an unfavorable state court "judgment" before ever being able to claim that he has been improperly denied admission.

In explaining that federal district courts have no jurisdiction to review decisions in this class of cases, the Court carefully distinguished between "general challenges to state bar admission rules and claims that a state court has unlawfully denied a particular applicant admission." Id. at 485, 103 S. Ct. at 1316. It is only in the latter class of cases, the Court explained, that an erroneous decision by the state court is, as it were, an element of the petitioner's claim of right. In contrast, a claimant can bring a general challenge to the rules promulgated by a state without first obtaining a "judgment" under those rules. Therefore, 28 U.S.C. § 1257 poses no barrier to a district court exercising jurisdiction over such a challenge.

Challenges to the constitutionality of state bar rules ... do not necessarily require a United States district court to review a final state-court judgment in a judicial proceeding. Instead, the district court may simply be asked to assess the validity of a rule promulgated in a judicial proceeding. If this is the case, the district court is not reviewing a state-court judicial decision. In this regard, 28 U.S.C. § 1257 does not act as a bar to the district court's consideration of the case and because the proceedings giving rise to the rule are nonjudicial the policies prohibiting United States district court review of final state-court judgments are not implicated. United States district courts, therefore, have subject-matter jurisdiction over general challenges to state bar rules, promulgated by state courts in nonjudicial proceedings, which do not require review of a final state-court judgment in a particular case. They do not have jurisdiction, however, over challenges to state-court decisions in particular cases arising out of judicial proceedings even if those challenges allege that the state court's action was unconstitutional. Review of those decision may be had only in this Court. 28 U.S.C. § 1257.

Id. at 486, 103 S. Ct. at 1316.

Thus, the impetus behind the Court's decision in Feldman is its identification of a class of cases in which res judicata will always constitute a defense to a party's attempt to bring a claim in federal district court, because the nature of the claim is such that it never arises except as the result of a judicial proceeding in a state court. As the Feldman Court explains, 28 U.S.C. § 1257 strips the federal district courts of jurisdiction over this unusual class of claims, depriving parties of the opportunity to bring their fruitless, always-already-barred claims in that forum.

Judge Rymer reads Feldman much more broadly. She argues that, under Rooker-Feldman, the California Supreme Court's decision in Eu prevents the lower federal courts from ever reviewing the constitutionality of Proposition 140. Since a decision by a federal court on the constitutionality of a state statute would "effectively ... 'overrule' " the merits of a decision by a state court regarding the same statute, Judge Rymer argues that the federal courts are disabled from ever reviewing the statute once a state court has rendered a final decision on the matter. See supra, Rymer concurrence at 857. So long as "the constitutional challenge raised in the federal court was 'inextricably intertwined' with the state court's decision," id., Judge Rymer would hold that state courts can permanently foreclose review of state statutes in lower federal courts.

Where Judge Rymer goes seriously wrong is in losing sight of the importance of the identity of parties to the application of the Rooker-Feldman doctrine. As the Third Circuit has explained, "Rooker-Feldman does not bar individual constitutional claims by persons not parties to earlier state court litigation." Valenti v. Mitchell, 962 F.2d 288, 298 (3d Cir. 1992); see also Dubinka, 23 F.3d at 221 (Under Rooker-Feldman analysis, "we must determine whether [a] constitutional claim [is] 'inextricably intertwined' with [a] state court's rulings in a particular plaintiff's case.") (emphasis added). " [T]he interests served by Rooker-Feldman are quite similar to those served by giving a state court judgment res judicata effect in a subsequent federal proceeding." E.B. v. Verniero, 119 F.3d 1077, 1091 (3rd Cir. 1997). Thus, the identity of the parties in similar state and federal proceedings is vital to determining what effect, if any, the Rooker-Feldman doctrine will have upon the jurisdiction of a federal district court. Where A and B bring a challenge to the same state statute in the same federal court proceeding, and A, but not B, has already brought a similar challenge against the same party in state court, then Rooker-Feldman may deprive lower federal courts of jurisdiction to hear A's claim, but it will have no impact on their jurisdiction to hear B's claim. This was precisely the result that obtained in Verniero, in which the Third Circuit was called upon by two different sets of litigants to review the constitutionality of "Megan's law," a compulsory notification law for sex-offenders. That court found that Rooker-Feldman deprived the federal courts of jurisdiction over the claim of one litigant, who had already obtained a judgment on the same claim in state court, but had no impact on the claims of other litigants who had not previously brought suit.

There can be no doubt that a federal court's pronouncement on the constitutionality of a state statute will have a serious impact on the vitality of previous state-court judgments concerning the same issue. But a litigant cannot invoke the Rooker-Feldman doctrine to prevent a federal court from "reviewing the merits" of a prior state judicial proceeding unless (at a minimum) he and his opponent were both parties to that proceeding, just as a litigant cannot raise a defense of res judicata by interposing an earlier judgment unless (at a minimum) his opponent is personally bound by that judgment. It is true that state and federal courts sometimes reach contrary conclusions as to the constitutionality of state statutes. Such clashes are unfortunate, and, as Judge Rymer rightly points out, it is sometimes appropriate for federal courts to abstain from deciding such issues in order to give states a chance to interpret their own statutes and thereby possibly to avoid constitutional problems. See supra, Rymer concurrence at 857 n.1; Railroad Comm'n of Texas v. Pullman Co., 312 U.S. 496, 61 S. Ct. 643, 85 L. Ed. 971 (1941). But the Rooker-Feldman doctrine, like the principle of res judicata, is concerned only with preserving the integrity of individual state court judgments, binding upon particular parties, from improper review by lower federal courts. Judge Rymer's attempt to transform this extremely narrow doctrine into an instrument for restructuring the entire relationship between state and federal courts is both unfortunate and misguided.

Given the importance of the right of the citizenry of a state to decide the form that its government will take (within the constraints imposed by the Federal Constitution in Article IV section 4, the Fourteenth Amendment, and elsewhere), I should not think it a remarkable proposition that, when the citizenry is asked to bring about profound changes in that government through a popular referendum, it must have adequate notice of what those profound changes will be. I do not suggest that such a vote may be invalidated simply upon proof that the voters do not understand, or have not made the effort to understand, the meaning of the initiative measure upon which they cast their votes.⁴  I do suggest that such a vote must be invalidated when the state does not provide information sufficient to allow any voter to know what it is that he is voting for.

Judge Rymer makes an important point. The initiative in this case deals with an issue of critical importance to the voters of the state: who will govern, citizen legislators or career legislators. While we may question the wisdom of submitting to the initiative process matters that might be better suited to the regular legislative process that is contemplated by traditional notions of republican government, this issue is, quintessentially, one that should be decided by a citizen vote (restricted, of course, by the demands of the Constitution). It is precisely because the issue before us is of such importance that we must ask whether the voters were given adequate notice of the essentials of the measure. It is in this context that the three-judge panel majority paid close attention to the information furnished by the state.

Several aspects of this litigation have tended to obscure the thoughtful attention that the important issues presented here should receive. Timing is the enemy of us all. Unfortunately, the political ramifications of this case have brought about unseemly haste, to the detriment of the deliberative process. In addition, the State of California has fundamentally shifted its position between its appearances before the California Supreme Court and before our court, leaving it in an awkward position, to say the least. Before the Supreme Court of California, the State vigorously argued that the initiative did not impose lifetime bans and urged the Supreme Court of California to ensure the constitutionality of the initiative by construing it to impose restrictions only on consecutive terms. The opponents of the initiative asserted that it was unconstitutional because it imposed a lifetime ban.

Now, before the federal courts, the State of California asserts two propositions: First, that the drafters of the initiative intended that it impose lifetime bans; but second, that a certain degree of ambiguity about the matter is not disqualifying and indeed may be injected purposefully to garner broader support from the voters. It is in this context that I respond to the majority's cursory dismissal of the important issue of the adequacy of the notice to the voters in this case, and to Judge O'Scannlain's suggestion that voters need not be given the information necessary to understand the implication of their votes.

The majority holds that the voters of California received adequate notice that, by voting in favor of Proposition 140, they were voting in favor of a lifetime ban. The majority opinion in the decision by the three-judge panel of this court more fully explains my opposition to that position. See Bates, 127 F.3d at 856-63. However, I add the following comments specifically in response to the majority's and Judge O'Scannlain's analyses.

The majority opinion is incorrect in its assertion that its holding, that the voters received adequate notice of the effect of Proposition 140, is "consistent with the California Supreme Court." Despite Judge O'Scannlain's implication to the contrary, the California Supreme Court did not rule on whether or not the voters received adequate notice; rather, the court answered only "the interpretive question whether Proposition 140 imposes a 'lifetime ban'." Eu, 286 Cal. Rptr. 283, 816 P.2d at 1314. The task of determining whether or not the voters received notice is fundamentally different from the task of interpreting the meaning of a voter initiative; in the latter context, the California Supreme Court acknowledged that it was concerned only with what the "average" voter "likely" believed. Id. at 1316. Not surprisingly, in making that assessment, the California Supreme Court explicitly admitted that "Proposition 140 is ambiguous as to its intent to impose a lifetime ban." Id. at 1315. It made this statement notwithstanding the State of California's contrary assertion that Proposition 140 "by its terms ... only limits the number of successive terms which incumbents in the elective offices may serve." Respondents' Brief in Legislature v. Eu ["Eu Brief"] at 4. Surely the adequacy of notice cannot turn on what a person of average intelligence "likely" understood regarding an admittedly ambiguous initiative; notice regarding the effect of one's vote requires clarity, see, e.g., United States v. Bass, 404 U.S. 336, 349, 92 S. Ct. 515, 523, 30 L. Ed. 2d 488 (1971); "careful and purposeful consideration," Greene v. McElroy, 360 U.S. 474, 507, 79 S. Ct. 1400, 1419, 3 L. Ed. 2d 1377 (1959); or "considered evaluation," Hampton v. Mow Sun Wong, 426 U.S. 88, 115, 96 S. Ct. 1895, 1910, 48 L. Ed. 2d 495 (1976).

The majority also inappropriately looks to the twenty-second amendment to the United States Constitution for support of its position. The twenty-second amendment states that " [n]o person shall be elected to the office of the President more than twice...." U.S. Const. amend. XXII (emphasis added). By contrast, Proposition 140 stated, " [n]o Senator may serve more than 2 terms," and " [n]o member of the Assembly may serve more than two terms." (emphasis added). As the state itself pointed out, this distinction has a specific meaning:

This verbal distinction between restrictions on the incumbent office-holder and restrictions on the individual person who at one time holds the office may be found in the constitutions of 18 states which impose term limitations upon their Governors. In each constitution, a restriction on the incumbent while he is an incumbent uses the title of the office ... while a restriction on the individual which extends after he leaves the office uses the word "person." ... [The state] knows of no constitution in which the title of an office is used to designate a person after he no longer holds that office.

Eu Brief at 23-24 n.11. Thus, the wording of the twenty-second amendment actually supports the opposite position from that advocated by the majority today. At the very least, it creates a serious question as to the intended meaning of Proposition 140, despite the majority's position that the meaning was clear to the voter.

The majority's reliance on the statements found in the materials distributed by those opposing Proposition 140 is also puzzling. While the opposition did suggest that the enactment of Proposition 140 would result in a lifetime ban, the state and those in favor of Proposition 140 never responded to that allegation. The State in fact never stated to the voters that a lifetime ban was the intent or the effect of the measure. As the state acknowledged, "this is not particularly surprising" since it allowed the supporters of Proposition 140 to "benefit from the ambiguity." Eu Brief at 21. Indeed, the state in oral argument before the en banc court went so far as to suggest that the voters have a right to vote for ambiguous legislation. Given the state's explicit admission that the state benefits by purposefully refusing to clarify the meaning and effect of an initiative on the ballot, I am amazed that the majority could find that the voters received adequate notice that by voting in favor of Proposition 140, they knew that they were voting to enact a lifetime ban.

Finally, the "surrounding circumstances" did not provide notice to the voters that they were enacting a lifetime ban. The majority relies upon the fact that the voters had two initiatives to choose from on the ballot--Proposition 140, and Proposition 131. Proposition 131, among other things, imposed consecutive term limit bans. The majority concludes from this fact that voters therefore must have known that because Proposition 131 involved consecutive term limit bans, Proposition 140 involved lifetime bans. I respectfully disagree. As the state itself argued, " [i]t simply cannot be said that the voters rejected Proposition 131 solely because the term limitation provisions of that measure" imposed a consecutive rather than a lifetime ban. Eu Brief at 25 n.13. Proposition 131 "covered several, complex issues" and was different from Proposition 140 in significant ways concerning funding and revenues. While editorials and the opponents' materials compared the effect of the two propositions in terms of the severity of the term limitations imposed by each, it cannot be said that such materials constitute "notice," particularly in the face of not a single reference to a lifetime ban in the text of the amendment, the proponents' ballot arguments, or the official statement prepared by the state.⁵ 

The evidence upon which the majority relies suggests at best that some of those who voted in favor of or against Proposition 140 believed that by doing so, they were voting in favor of or against a lifetime ban.⁶  It necessarily follows, however, that others who voted in favor of or against the proposition had no notice that they were voting in favor of or against a lifetime ban and so had no notice that they were voting on a severe limitation to their fundamental right to vote for candidates of their choice. Reynolds, 377 U.S. at 555, 84 S. Ct. at 1378. We ask little of the state when we request that it give voters adequate notice of the effect of their vote, particularly when the state asks voters to restrict their own fundamental rights. In contrast, the state loses little when we take away the advantage it gains by fostering an ambiguity as to the effect of a voter initiative.

As Judge Rymer's concurrence appropriately emphasizes, "decisions about how people are to participate in their own representative government and how they choose to represent themselves is 'the very essence of all self-government.' " See supra, Rymer concurrence at 15118 (quoting Laurence H. Tribe, American Constitutional Law 398 (2d ed.1988)) (emphasis in original). Ironically, however, the majority would have the voters of California make those decisions while having to guess at the likely outcome of their votes, having to rely upon newspaper editorials and materials circulated by the opponents of the voter initiatives, rather than upon the plain language of the initiative itself and the explanatory materials produced by the state. It is difficult to determine how the right to vote--particularly the right to vote on issues affecting one's own fundamental rights--can be meaningful when the state is allowed to force people to vote at their peril rather than add a few simple words to clarify the position for or against which they are being asked to cast their ballots.

The majority disposes of the merits of petitioner's constitutional challenge to lifetime term limits in an opinion that barely spans one page of the federal reporter. The majority apparently believes that plaintiffs' challenge presents an easy question. I find the question to be an extremely difficult one.

When a court is called upon to analyze the constitutionality of restrictive qualifications imposed by a state upon candidates for office, it faces a dilemma to which the Supreme Court has not yet offered any answer: How is the court to determine whether a qualification is a "proper" one or not? It seems clear that some qualifications are eminently "proper" and should not require any extraordinary justification by a State in order to survive scrutiny. A requirement that candidates be adults, for example, or that they actually reside in the districts that they seek to represent, fits with our understanding of the most basic of requirements in a representative democracy: That candidates for office be citizens who are willing and able to represent the interests of a particular community. Any constitutional analysis that would require a State to offer a highly particularized explanation for imposing basic age and residency requirements upon candidates for elective office would be extraordinarily invasive.

Even so, the Supreme Court has at least suggested that such an invasive analysis may be called for. On a number of occasions, the Court has emphasized the "fundamental principle of our representative democracy ... 'that the people should choose whom they please to govern them.' " Powell v. McCormack, 395 U.S. 486, 547, 89 S. Ct. 1944, 1977, 23 L. Ed. 2d 491 (1969). "The people are the best judges who ought to represent them," the Court has written, and " [t]o dictate and control them, to tell them whom they shall not elect, is to abridge their natural rights." U.S. Term Limits v. Thornton, 514 U.S. 779, 794-95, 115 S. Ct. 1842, 1850-51, 131 L. Ed. 2d 881 (1995) (quoting 2 Elliot's Debates 292-93 (statement of Robert Livingston)). Indeed, the Court has gone so far as to say that " [t]he right to vote freely for the candidate of one's choice is of the essence of a democratic society, and any restrictions on that right strike at the heart of representative government." Reynolds v. Sims, 377 U.S. 533, 555, 84 S. Ct. 1362, 1378, 12 L. Ed. 2d 506 (1964) (emphasis added). Taken literally, these ringing phrases would suggest that every attempt by a state to set restrictive qualifications concerning who can and who cannot serve as a candidate for elective office implicates rights that are "fundamental" and requires careful scrutiny by a reviewing court.

It will not do to respond that the Court has only meant to secure to voters the right to vote for the qualified candidate of their choice, for this merely begs the question, what qualifications may a State impose? By narrowing the field of candidates who are eligible to run for office, lifetime term limits might permanently deprive an entire legislative district of the opportunity to vote for the otherwise-"qualified" candidate of its choice, even if that district had overwhelmingly opposed the imposition of term limits in the first place. Ascertaining the constitutional principles that should govern such a dilemma is of the highest importance. The qualifications for federal legislative office are absolutely fixed in the Constitution. See U.S. Const., Art. I § 2 cl. 2-3; Thornton, 514 U.S. at 837-38, 115 S. Ct. at 1871-72. There are no such clear constitutional guideposts for state legislators. Thus, unless the Court means for lower federal courts to take literally its stirring pronouncements concerning " [t]he right to vote freely for the candidate of one's choice," Reynolds, 377 U.S. at 555, 84 S. Ct. at 1378 these repeated, powerful statements do not aid in answering the basic question that is before us: How is the court to determine whether a qualification is a "proper" one or not?

The majority rests its conclusion that states may freely impose lifetime term limits upon state elected officials primarily upon its assertion that "term limits on state officeholders is a neutral candidacy qualification," one that " [does] not constitute a discriminatory restriction." Majority opinion at 846-847. The majority looks to the Supreme Court's cases involving restrictions on candidate access to voting ballots for the proposition that "the State's important regulatory interests are generally sufficient to justify" such restrictions, provided that they are "reasonable" and "nondiscriminatory." Burdick v. Takushi, 504 U.S. 428, 434, 112 S. Ct. 2059, 2063, 119 L. Ed. 2d 245 (1992). The majority concludes that Proposition 140 does not "limit [ ] political participation by an identifiable political group whose members share a particular viewpoint, associational preference, or economic status," Majority opinion at 847 (quoting Anderson v. Celebrezze, 460 U.S. 780, 789, 103 S. Ct. 1564, 1570, 75 L. Ed. 2d 547 (1983)), and hence that lifetime term limits do not provoke any searching judicial scrutiny.⁷ 

I find the majority's analysis problematic, at best. In my view, U.S. Term Limits v. Thornton forecloses the analytical approach that the majority invokes. As the majority points out, Thornton involved a challenge to an attempt by the State of Arkansas to impose term limits on members of the U.S. Congress. That attempt raised serious concerns under the Qualifications Clauses, U.S. Const., Art. I § 2 cl. 2-3, concerns which are not implicated here, and the Court rested its holding in Thornton primarily upon its analysis of those particular concerns. During the course of its opinion, however, the Court also examined the Arkansas amendment in light of its ballot-access line of cases, finding that there was nothing in those cases that could save the challenged amendment. See Thornton, 514 U.S. at 834-35, 115 S. Ct. at 1869-70. That analysis is entirely at odds with the approach that the majority has adopted in this case.

The State of Arkansas argued in Thornton that its attempt to impose term limits on members of Congress was in fact nothing more than a permissible exercise of its power, under the Elections Clause of the Federal Constitution, to regulate the "Times, Places and Manner" of federal elections. U.S. Const., Art. 1 § 4 cl. 1; see Thornton, 514 U.S. at 832, 115 S. Ct. at 1868. In rejecting that argument, the Court looked to its ballot-access cases and explained that the types of regulatory procedures that those cases involved (many of which constituted permissible regulations under the Elections Clause) were qualitatively different from the imposition of a qualification for candidacy in the form of a lifetime term limitation. As the Court explained, " [t]he provisions at issue in [the ballot-access / Elections Clause cases] were ... constitutional because they regulated election procedures and did not even arguably impose any substantive qualification rendering a class of potential candidates ineligible for ballot position." Thornton, 514 U.S. at 835, 115 S. Ct. at 1870.

Once again, the imposition of qualifications upon state legislators does not implicate the interplay between the Qualifications Clause and the Elections Clause that the Court was primarily concerned with in Thornton. But the Court was quite clear in Thornton that its conclusion that the ballot-access cases could not save the Arkansas amendment relied both upon the nature and extent of the burden imposed by lifetime limits and upon the unique limitations embodied in the Qualifications Clause. Thus, the Court explained that its ballot-access cases "provide little support for the contention that a state-imposed ballot access restriction is constitutional when it is undertaken for the twin goals of disadvantaging a particular class of candidates and evading the dictates of the Qualifications Clause." Id. at 835, 115 S. Ct. at 1870 (emphasis added). As to the former consideration, the Court explained that the types of ballot regulations that the Court has upheld in cases like Burdick are qualitatively different from "measures that exclude candidates from the ballot without reference to the candidates' support in the electoral process." Id.

The Court's analysis of its ballot-access cases in Thornton does not by itself require a finding that lifetime term limits on state legislators are unconstitutional. It does compel the conclusion, however, that such limits "disadvantag [e] a particular class of candidates." Id. There is no difference between the operation of state and federal term limits in this regard. Both "exclude candidates from the ballot without reference to the candidates' support in the electoral process." Id. The clear import of the Court's analysis in Thornton is that lifetime term limits cannot enjoy the deferential review that the ballot-access cases reserve to those procedural reforms that "impose only reasonable, nondiscriminatory restrictions upon the First and Fourteenth Amendment rights of voters." Burdick, 504 U.S. at 434, 112 S. Ct. at 2063. Indeed, Thornton could be read for the broad proposition that the ballot-access cases are simply inapplicable to qualification regulations, of which lifetime term limits are an example. See supra, Rymer concurrence at 852. It at least stands for the narrower proposition that, even under the ballot-access cases, lifetime term limits have the effect of "disadvantaging a particular class of candidates," Thornton, 514 U.S. at 835, 115 S. Ct. at 1870 and hence provoke searching judicial review. See Anderson, 460 U.S. at 793, 103 S. Ct. at 1572. Under either interpretation, Thornton forecloses the deferential approach that the majority has adopted in this case.

Before going on to discuss the constitutionality of Proposition 140 under the stricter standard that Thornton requires us to employ, I think it appropriate to explore further the Supreme Court's conclusion that lifetime term limits discriminate against "a particular class of candidates." Thornton, 514 U.S. at 835, 115 S. Ct. at 1870. The majority in Thornton apparently took this to be a self-evident proposition, but, as the district court in the present case aptly observed, "the terms 'reasonable' and 'nondiscriminatory' [and their converse] are not self-defining." Bates v. Jones, 958 F. Supp. 1446, 1462 (N.D. Cal. 1997). Justice Kennedy took up this question in Thornton at somewhat greater length than did the majority, explaining that lifetime term limits had the dubious effect of burdening the rights of voters⁸  based on the manner in which they had exercised the franchise in past elections.

Not the least of the incongruities in the position advanced by Arkansas is the proposition, necessary to its case, that it can burden the rights of resident voters in federal elections by reason of the manner in which they earlier had exercised it. If the majority of the voters had been successful in selecting a candidate, they would be penalized from exercising that same right in the future.

Thornton, 514 U.S. at 844, 115 S. Ct. at 1874 (Kennedy, J., concurring). Justice Kennedy made this observation during the course of a discussion about the burden Arkansas had placed upon federal voting rights, but he also cited Anderson and suggested that the restriction raised "First Amendment concerns." Id. Indeed, the concern that Justice Kennedy expressed over a restriction on the right to vote or hold office that looks to "the manner in which [a voter or candidate] earlier had exercised" that right, id., echoes the Thornton majority's suspicion of "measures that exclude candidates from the ballot without reference to the candidates' support in the electoral process." Id. at 835, 115 S. Ct. at 1870.

In its ballot-access cases, the Court has admonished federal courts to "examine in a realistic light the extent and nature of [candidate restrictions'] impact on voters." Anderson, 460 U.S. at 786, 103 S. Ct. at 1568 (quotation omitted). On the most basic level, as the district court in the present case observed, "Proposition 140 prevents [California voters] from expressing their political preference for candidates with legislative experience." Bates, 958 F. Supp. at 1463. "California voters who value legislative experience ... are not simply required to defer expressing their preference, they are forever prevented from expressing their preference at the ballot box." Id. The district court found that this impact alone was enough to warrant strict scrutiny under Anderson and Burdick, a conclusion that was surely warranted in light of the Supreme Court's conclusion that lifetime term limits "disadvantag [e] a particular class of candidates." Thornton, 514 U.S. at 835, 115 S. Ct. at 1870.

Even so, this explanation, without more, is not wholly satisfactory. The implicit rejoinder of the majority's analysis is an attractive one, at least on its face. Term limits affect all candidates equally, without regard to the candidates' parties, their political affiliations, or their resources. Thus, on their face, term limits do not seem to disadvantage any particular class of voters, unless we can say that there are voters who value legislative experience, not merely for the incumbency-based benefits that it can confer upon fortunate districts, but as a political and ideological reason for selecting a representative. I do not doubt that there are some voters who value legislative experience for precisely this reason. It is not self-evident, however, that such voters constitute a "particular class." There surely are at least some voters who value candidates, on an ideological level, for reasons that would be implicated by the ballot-access restrictions that the Court has upheld in the past. I can easily imagine, for example, that there are voters (albeit few) who might have principled reasons for trusting only a candidate who refuses to take part in any organized balloting and primary activities, instead insisting that he be elected, if at all, by popular acclaim through the write-in process. Cf. Burdick, 504 U.S. at 435-36, 112 S. Ct. at 2064-65. How, then, is a court to decide, in a close case, when a restriction has the disfavored effect of "disadvantaging a particular class of candidates?" Thornton, 514 U.S. at 835, 115 S. Ct. at 1870.

I am persuaded that the animating force behind the Court's conclusion in Thornton that term limits have this effect is its observation that term limits "exclude candidates from the ballot without reference to the candidates' support in the electoral process." Id. (emphasis added). The regulations that the Court has upheld in its ballot-access cases have been aimed at "seeking to assure that elections are operated equitably and efficiently," Burdick, 504 U.S. at 433, 112 S. Ct. at 2063 and at "protect [ing] the integrity and reliability of the electoral process itself," Anderson, 460 U.S. at 788 n. 9, 103 S. Ct. at 1570 n.9 even though they might, in theory, result in some voter's ideologically preferred candidate being excluded. What those restrictions have in common is that they have asked questions that relate to a candidate's level of engagement with the electoral process. Has the candidate made use of readily available methods of gaining access to the ballot? See Burdick, 504 U.S. at 435-36, 112 S. Ct. at 2064-65. Has the candidate demonstrated a threshold level of popular support? See Munro v. Socialist Workers Party, 479 U.S. 189, 192-93, 107 S. Ct. 533, 535-36, 93 L. Ed. 2d 499 (1986). Has the candidate identified himself with a particular party for purposes of a given election? See Storer v. Brown, 415 U.S. 724, 726-28, 94 S. Ct. 1274, 1277-78, 39 L. Ed. 2d 714. It is not enough, of course, for a candidate restriction to fit this general description. If the restriction burdens the rights of voters in too disparate or severe a fashion, it still may not pass muster. See, e.g., Anderson, 460 U.S. at 806, 103 S. Ct. at 1579. But this is the type of question that the ballot-access cases suggest that states ought to be asking in passing candidate restrictions. The candidate qualifications that I discussed at the beginning of this section--age and residency--would likewise merit deference under this approach. Clearly, residency requirements are directly concerned with the level of a candidate's engagement with a particular community. Similarly, people below a certain age do not enjoy all the rights and responsibilities of citizenship, and reasonable age requirements help to ensure that a candidate can fully appreciate, and effectively represent, the most basic interests of his constituency.

What the Court suggested in Thornton, however, was that a "measure [ ] that exclude [s] candidates from the ballot without reference to the candidates' support in the electoral process," Thornton, 514 U.S. at 835, 115 S. Ct. at 1870 is asking the wrong type of question, at least as far as the ballot-access cases are concerned. A restriction that imposes a "substantive qualification rendering a class of potential candidates ineligible for ballot position," id., always carries with it the danger that particular classes of voters will be disadvantaged, perhaps in ways that will be difficult to foresee. The ballot-access cases, Thornton suggests, represent a compromise under which states are permitted to pass restrictions aimed at protecting the electoral process, but only so long as those restrictions limit themselves to asking about a candidate's level of engagement and support in that process itself. See id. If the burdens that the restriction imposes are not too severe, then the restriction will enjoy a presumption of validity. If the restriction disadvantages candidates for reasons having nothing to do with a lack of engagement and support in the electoral process, however, then that presumption of validity disappears. At that point, the "fundamental principle ... 'that the people should choose whom they please to govern them,' " Powell v. McCormack, 395 U.S. 486, 547, 89 S. Ct. 1944, 1977, 23 L. Ed. 2d 491 (1969), reasserts itself and requires courts to look closely at the challenged restriction. Such is the case with lifetime term limits, which prevent candidates from running for office precisely because they are engaged with the political process and have found considerable support in it. See Thornton, 514 U.S. at 844, 115 S. Ct. at 1874 (Kennedy, J., concurring). We cannot accord such a regulation a casual presumption of validity. Rather, we must weigh it carefully against the interests that the State asserts in its defense.

I recognize the uniquely compelling interest that states enjoy in experimenting with new forms of republican government under our federal system. Judge Rymer provides an account of that interest in her opinion that is powerful. See supra, Rymer concurrence at 851-852. As Justice Brandeis has famously written, " [i]t is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country." New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S. Ct. 371, 386, 76 L. Ed. 747 (1932) (Brandeis, J., dissenting). As Judge Rymer explains, this principle is applicable to political reform as well, for, " [t]hrough the structure of its government, and the character of those who exercise government authority, a State defines itself as a sovereign." Gregory v. Ashcroft, 501 U.S. 452, 460, 111 S. Ct. 2395, 2400, 115 L. Ed. 2d 410 (1991); see supra, Rymer concurrence at 852.

In this regard, a limitation on the number of terms that a candidate may serve in state government implicates federalism interests that pull in the opposite direction from the federalism interests implicated by a similar restriction on elected federal officials. As the Court explained in Thornton, it was of central concern to the Framers of our Constitution that the States enjoy a relationship to the federal government that would not vary from state to state in order that they might be unified into a single, stable Republic. See Thornton, 514 U.S. at 806-07, 115 S. Ct. at 1856-57. "Permitting individual States to formulate diverse qualifications for their representatives would result in a patchwork of state qualifications, undermining the uniformity and the national character that the Framers envisioned and sought to ensure." Id. at 822, 115 S. Ct. at 1864. Rather, it is the bedrock of stability that our uniform federal government offers that in turn makes it possible for States to experiment safely with novel approaches to intractable social, economic and political problems. The principle of federalism thus weighs heavily in favor of allowing California to experiment with a government of "citizen legislators" like the one that Proposition 140 envisions.

It is necessary, however, to articulate some limitations on the ability of States to alter their structure of government on the strength of these federalism interests. Unfortunately, the ordinary paradigm of strict scrutiny analysis is not helpful in this regard. Once it is acknowledged that the States have a compelling interest in experimenting with new forms of government, the concept of "narrow tailoring" has limited application, for any novel governmental structure is, by definition, narrowly tailored to the goal of experimenting with that particular governmental structure. But the States do not have a completely free hand in choosing how to organize themselves. That is why the question before us has no easy solution.

It is clear, at the very least, that a State cannot discriminate in an invidious fashion against its citizens, even in the name of sovereignty and federalism. See Ashcroft, 501 U.S. at 460, 111 S. Ct. at 2400. Disadvantaging candidates on the basis of their political ideology would unquestionably constitute such "invidious discrimination." I have said that I am not convinced that favoring candidates on the basis of legislative experience constitutes a sufficiently distinct "political ideology" to warrant the invalidation of term limits as an invidious candidate qualification. I consider the question a close one, however. Several members of our court for various reasons would not reach the issue in this case, nor would I, in light of my view that the process that produced Proposition 140 was infirm. It is a question that, ultimately, the Supreme Court must address, and its decision will be a weighty one. Where along the fulcrum the balance lies is not easy of resolution.

I take the occasion here, however, to express my further discomfort with any ruling that would permanently foreclose a court from concluding, after seeing term limits in action, that they were wrong: that a preference for legislative experience does, in fact, constitute a genuine and distinct voter preference or political ideology, and that lifetime term limits have deprived an identifiable group of voters of the opportunity to elect representatives who will represent their interests and promote their political ideals.

When "there is a serious factual controversy over the feasibility of recognizing [a] claimed right without at the same time making it impossible for the State to engage in an undoubtedly legitimate exercise of power," Washington v. Glucksberg, --- U.S. ----, ----, 117 S. Ct. 2258, 2292, 138 L. Ed. 2d 772 (1997) (Souter, J., concurring), it may sometimes be appropriate for federal courts to deny a claim of right while remaining open to reaching a different conclusion if a demonstrable infringement on individual liberties can later be shown. This principle holds especially true when the organization of our federal system is at issue, since one of the very purposes of that system is to foster investigation and experimentation by state legislatures. See id. at ----, 117 S. Ct. at 2293 (Souter, J., concurring); Glucksberg, --- U.S. at ----, 117 S. Ct. at 2303 (O'Connor, J., concurring). At a minimum, I suggest that the majority should recognize that what it does today is to validate an experiment that may not prove out. If experience demonstrates that the predicate that underlies the majority's decision is incorrect, and that lifetime term limits consistently disadvantage a particular class of voters on the basis of an articulable political ideology, then our court should remain open to revisiting its holding.* * *

I voice my strong dissent from the majority's view that the voters of California received adequate notice from the State regarding the true import of their vote on Proposition 140. From reading their opinion, one might conclude also that the majority did not appreciate either the gravity or the closeness of the issues upon which they were called to deliberate.